CN103403543A - Colon cancer gene expression signatures and methods of use - Google Patents

Abstract

A gene expression signature of colon cancer, microarrays including them and methods of using the colon gene expression signature are provided. The gene expression signature is especially useful for determining the prognosis of a patient diagnosed with colon cancer, such as stage II colon cancer. The gene signature described herein is also useful for determining effectiveness of surgical resection with or without adjuvant chemotherapy, and determining possibility of cancer recurrence in patients with colon cancer.

Description

Colon cancer gene expression signature and using method

The cross reference of related application

The application requires the rights and interests of the U.S. Provisional Application of submitting on January 25th, 2011 number 61/435,922, its this by reference integral body incorporate into.

Invention field

Present disclosure relates to colon, for example the gene expression profile in colon cancer tissue.Particularly, present disclosure relates to the sensitive method of measuring mRNA level in the biopsy colon tumor tissue, and described biopsy colon tumor tissue comprises the paraffin-embedded biopsy material that files.In addition, present disclosure provides the array of the gene expression signature that is formed for prognosis, diagnosis and treatment colon cancer to express transcript.

Background of invention

About 30% of all colorectal cancer patients are diagnosed with II phase disease (people such as Jemal, CA Cancer J.Clin., 2004).Survivals in 5 years of the II phase colorectal cancer patients by operative treatment are about 75-80%, proves that most of patient passes through independent surgical healing (Benson, The Oncologist, 2006; The people such as Nauta, Arch.Surg., 1989.).Yet about 20-25% of these patients will develop recurrence disease (Benson, The Oncologist, 2006 within its life cycle; The people such as Gill, J.Clin.Oncol., 2004).In theory, these patients should benefit from NACT.Yet only approximately 3-4% patient uses survivals in 5 years of NACT to have absolute improvement the (Benson, The Oncologist, 2006 in II phase colon cancer; The people such as Andr é, Annals of Surgical Oncology2006).Therefore, these patients of American Society of Clinical Oncology's guideline recommendation should not use NACT conventional therapy (people such as Benson, J.Clin.Oncol., 2004).However, but be clear that, about 20% the II phase colorectal cancer patients that is in higher risk of recurrence may be candidate (Benson, The Oncologist, 2006 of supplemental treatment; The people such as Nauta, Arch.Surg., 1989; The people such as Gill, J.Clin.Oncol., 2004; The people such as Andr é, Annals of Surgical Oncology2006.).

In the disease such as colon cancer, the first treatment is normally most important, and maximum chance of success is provided, and therefore, exists to use for the moment of patient's colon cancer and the most effectively treats the demand of as first, treating.This is being impossible traditionally always, because do not have method to can be used to predict what drug therapy, will be the most effective for particular individual physiology.Many times, the patient does not need to experience the drug toxicity therapy.For example, in II phase tumour is carried down and moved (TNM) colon cancer, also do not have which patient of determining can respond afterwards in operation the method for NACT.20% the II phase patient who is in risk of recurrence after operation only 1/3rd from chemotherapy any benefit of acquisition.This means that carrying out NACT makes some patients be exposed to unnecessary treatment.Perhaps, the decision of not carrying out NACT in this stage can make some patients be exposed to higher cancer return risk.

At present, the diagnostic test of using in clinical practice is based on single analyte testing, and therefore do not catch the potential value of understanding tens of kinds of unlike signal thing Relations Amongs.And diagnostic test is not usually quantitative, and this depends on immunohistochemistry.The laboratory of the method through being everlasting different produces different results, and part is not standardized because of reagent, and partly because explain it may is subjective, and may be not easy quantitatively.Test based on RNA is also often used, because RNA degradation problem in time obtains the flesh tissue sample for the fact of analyzing with being difficult to from the patient.Fixing paraffin-embedded tissue more easily obtains, and has set up the method that detects RNA in fixing organization.Yet these methods do not allow usually from a small amount of investigation of materials lots of genes (DNA or RNA).Therefore, traditionally, except detecting, the immunohistochemistry that is used for protein also seldom uses fixing organization.

Recently, several groups announced about by the microarray gene expression analysis to the research of various cancer types classification (referring to for example, the people such as Golub, Science286:531537 (1999); The people such as Bhattacharjae, Proc.Natl.Acad.Sci.USA98:1379013795 (2001); The people such as Chen-Hsiang, Bioinformatics17 (supplementary issue .1): S316S322 (2001); The people such as Ramaswamy, Proc.Natl.Acad.Sci.USA98:1514915154 (2001), the people such as Salazar, Journal of Clinical Oncology29:17-24 (2010), the people such as O ' Conneell, the people such as Journal of Clinical Oncology28:3937-3944 (2010) and Kerr, Journal of Clinical Oncology27 (supplementary issue) 15s (2009)).Yet these researchs mainly concentrate on the classification that improves and refines the various cancer types of having set up, and the general neodoxy that does not provide the relation of difference expression gene, and not related discovery result and therapeutic strategy are to improve the clinical effectiveness of cancer therapy.In addition, still based on the validity of new reactive compound but not the integrated approach of pharmacogenomics is proceeded treatment of cancer and colon cancer clinical trial, the integrated approach of described pharmacogenomics has utilized genomic constitution and the genotype of patient's tumour to set up personalized therapeutic scheme.

Although modern molecular biology and biological chemistry have disclosed its activity influence tumour cell behavior, its differentiation state and to the 100 kinds of genes that surpass of the susceptibility of some medicine or resistance, but there is a few exceptions, also for routine, do not make the identity of exploring these genes about the purpose of the clinical decision of drug therapy.

Brief summary of the invention

Need to identify the biomarker for the prognosis of predicting the patient who suffers from colon cancer.The patient is categorized as excessive risk (poor prognosis) or the ability of low-risk (good prognosis) can be selected suitable therapy for these patients.For example, high-risk patient may be benefited from active treatment, and therapy may not have significant advantage to the low-risk patient.Yet,, although this demand is arranged, also there is no this available way to solve the problem.

Therefore, need to be based on the prognosis technology of microarray, it provides about using particular treatment for the doctor, and the resection that for example is with or without chemotherapy recovers or recurs the information of possibility afterwards.Also need can Accurate Diagnosis colonic diseases, particularly diagnosing colon moment, perhaps can predict the technology of colonopathy patient to specific therapy response.Concrete understanding about tumour in the cancer patient will be very useful aspect prolongation alleviation, raising patients ' life quality and minimizing health care cost.This type of technology can also be used for patient's candidate of the clinical testing of screening novel therapeutic Compounds and methods for, to promote supervision, examines process.

Expression signature from the colon cancer that meets these demands is disclosed.Disclosed signature can be used for the application in colon cancer prognosis, diagnosis of colon cancer and patient's grouping.In some embodiments, these results allow separately or with the assessment of the genome evidence of the effect of the operation of the NACT combination that is used for the treatment of colon cancer.Signature described herein may be significant in distinguishing two diagnosis or prognosis result, and can distinguish two diagnosis or prognosis result.An importance of present disclosure is to mate patient and optimal treatment with the expression of some gene of measuring in colon cancer tissue, and prognosis information is provided.Therefore, the method for using this colon cancer signature is disclosed.Disclosed method comprises the expression that detects available from least two kinds of listed colon cancer associated nucleic acid molecules of table 6 in experimenter's the sample that comprises nucleic acid, and the expression of more described at least two kinds of colon cancer associated nucleic acid molecules or from the decision-making scoring and contrast threshold value of its derivation.Prediction as requested, the contrast threshold value can be indicated the diagnosis of colon cancer, the known classification of indication colon cancer, the known response of indication to treatment, indication has long-term surviving history, and the indication recurrence is historical etc.

In each embodiment, RNA separates from colon's sample, and for the preparation of gene expression profile.In relating to some embodiment of cancer prognosis, sample is the colorectum tumor specimen, for example the colon cancer sample.In certain embodiments, gene expression profile relates to the expression that detection table 6 is listed and can also list at least 50 transcripts of table 1 and/or table 2.The sum of the transcript that detects in gene expression profile can change.For example, in some embodiments, the sum of the transcript that detects in spectrum is approximately 200 to approximately 1000 or approximately 400 to approximately 800, and perhaps in other embodiments, the number of transcript is approximately 500 to approximately 700 or approximately 550 to approximately 650.In each embodiment, in table 6 listed at least about 50, at least about 100, at least about 200, at least about 300, at least about 400, at least about 500, at least about 600 or all transcripts detect as the part of the sum of transcript.When detecting extra transcript (except those of table 6), they can randomly be selected from signal or expression contrast, and are known transcript of expressing in colon cancer in some embodiments, for example by Colorectal Cancer DSA ^TMThose that measure.In certain embodiments, extra transcript can also be indicated the colon cancer prognosis.

In excessive risk and low-risk patient group, for example have in the patient of high or low clinical recurrence risk, sign to the scoring of patient's express spectra for the expression of the expression based on the listed transcript of table 6, and result can be used for determining therapeutic process.For example, the patient who is confirmed as high-risk patient can treat at the used after operation NACT.For the patient who is considered to the low-risk patient, NACT can not carry out after operation.Therefore, in some aspects, the invention provides the method for the preparation of the gene expression profile of the colon cancer tumour of indicating risk of recurrence.

Present disclosure also provides the method that is used for the prognosis colon cancer.According to the method for this aspect comprise preparation colon cancer sample ( For example, gene expression profile as described herein).Then gene expression profile is classified or mark for gene expression signature described herein.In each embodiment, gene expression signature is listed and expression that can also list at least 50 transcripts of table 1 and/or table 2 based on table 6.In some embodiments, signature based on transcript sum less than approximately 800, less than approximately 700, less than approximately 600, less than approximately 500, less than approximately 400, less than approximately 300, less than approximately 200 or less than about 100 transcripts, and it comprises the transcript from table 6.For example, signature can based on from table 6 at least about 400, at least about 500 or at least about the expression of 600 transcripts.Randomly, the transcript from table 6 comprises the transcript that table 1 is listed.

The method of preparation experimenter's personalized colon cancer genome spectrum is also disclosed.Described method comprises the expression that detects available from least two kinds of listed colon cancer associated nucleic acid molecules of table 6 in experimenter's the sample that comprises nucleic acid, and produces the report of having summarized the data that obtain by gene expression analysis.

In some examples of disclosure method, from the nucleic acid determination expression available from the experimenter, described nucleic acid comprises from RNA and/or the cDNA of the rna transcription of the colorectum tissue samples available from the experimenter (for example colon cancer sample) extraction.

Nucleic acid probe and primer (and this type of probe of array and primer) for detection of the gene expression signature of colon cancer are also disclosed.In some instances, probe is the part for detection of the array of colon cancer signature.

The following detailed description of several embodiments that aforementioned and other features of present disclosure and advantage will be carried out by reference to accompanying drawing and become more obvious.

The accompanying drawing summary

Fig. 1 provides and has shown that being used for deriving the colon cancer transcript expresses the process flow diagram of the exemplary process of signature.

Fig. 2 provides and has shown use Colorectal Cancer DSA ^TMCarry out the process flow diagram of the exemplary outline of the generation of II phase colon cancer prognosis signature and checking.

Fig. 3 A provides the figure of recipient's operating characteristics (ROC) curve of 636 transcript prognosis signatures in the training group.

Fig. 3 B provides the Kaplan-Meier figure from the recurrence of the training data of candidate family.

Fig. 4 A provides the figure of recipient's operating characteristics (ROC) curve of 636 transcript prognosis signatures in the checking group.

Fig. 4 B provides the Kaplan-Meier figure from the recurrence of the verification msg of candidate family.

Fig. 5 provides the Kaplan-Meier figure from the overall survival of the verification msg of candidate family.

Fig. 6 is table 3 as described below.

Fig. 7 is table 6 as described below.

The form summary

Table 1 provides the list of core colon 10 candidate's transcripts that signature comprises.Identified that these transcripts have maximum effect to sample classification for poor prognosis group of becoming reconciled.

Table 2 provides the list of colon 178 unique transcripts that signature comprises.This table comprises the direction of the transcript of expressing in the weight grade (weight rank) of 636 transcripts signature transcription things and colon.

Table 3 provides crucial patient and tumour feature in research to sign to identify 636 transcripts.

Table 4 provides the training group of cross validation and for the identification of the performance standard of the checking group of transcript signature.

Table 5 provides and has shown that the risk of patient age, Gender, pT stage, tumour grade, knub position and mucus/non-mucus hypotype state is than the statistic analysis result of (Hazards Ratio).

Table 6 provides the list of the transcript that 636 transcript colon signatures comprise.

Sequence table

Use the standard alphabet abbreviation as the nucleotide base of definition in 37C.F.R. § 1.822 to show listed nucleotide sequence and amino acid sequence in appended sequence table.The only chain that has shown each nucleotide sequence, but understand by to shown in any mentioning of chain comprise complementary strand.In appended sequence table:

SEQ ID NO:1-636 is the oligonucleotides transcript from human colon carcinoma.

Sequence table is submitted to the document form of called after ADL-0311_Sequence_Listing.txt as ASCII text file, and it created on January 25th, 2012, and was 232,154 bytes, and it is incorporated into by reference at this.

Describe in detail

I. terminology overview

Unless otherwise defined, Science and Technology term used herein have with disclosure one skilled in the art the identical implication usually understood.In molecular biology, the definition of common term can be referring to Benjamin Lewin, Genes IX, and Jones and Bartlet publishes, 2008 (ISBN0763752223); The people such as Kendrew (editor), The Encyclopedia of Molecular Biology, Blackwell Science Ltd. publishes, 1994 (ISBN0632021829); Robert A.Meyers (editor), Molecular Biology and Biotechnology:a Comprehensive Desk Reference, VCH Publishers, Inc. publishes, 1995 (ISBN9780471185710); The people such as Singleton, Dictionary of Microbiology and Molecular Biology the 2nd edition, J.Wiley﹠amp; Sons (New York, N.Y.1994), and March, Advanced Organic Chemistry Reactions, Mechanisms and Structure the 4th edition, John Wiley﹠amp; Sons (New York, N.Y.1992).

Unless context clearly indicates in addition, singular references " (a) ", " a kind of (an) " and " being somebody's turn to do (the) " comprise a plurality of denotion things.Similarly, unless context clearly indicate in addition, word "or" intention comprise " with ".Term " comprises (comprise) ", and expression " comprises (include) ".In situation about clashing, with this instructions (comprising terminological interpretation), be as the criterion.

, for the ease of each embodiment of summary present disclosure, provide following terminological interpretation:

Amplifier nucleic acid molecule: to increase nucleic acid molecules, for example gene or genetic fragment, the transcript that example is as shown in table 6, copy number.The product that obtains is called as amplified production.

The example of amplification in vitro is PCR (PCR).Other examples of Amplification Technologies comprise quantitative PCR in real time, strand displacement amplification (referring to U.S. Patent number 5,744,311); Without transcribing isothermal duplication (referring to U.S. Patent number 6,033,881); Repair chain reaction amplification (referring to International Patent Publication No. W WO90/01069); Ligase chain reaction amplification (referring to EP-A-320308); Gap-fill ligase chain reaction amplification (referring to U.S. Patent number 5,427,930); The ligase of coupling detects and PCR (referring to U.S. Patent number 6,027,889); And NASBA ^TMRNA is without transcription amplification (referring to U.S. Patent number 6,025,134).

Array: molecule, for example arrangement in biomacromolecule (for example nucleic acid molecules) or the addressable point of biological specimen (for example histotomy) in substrate or in substrate.In some instances, array is to be combined array with the polynucleotide probes that obviously do not removed in the crossover process probe of nucleotide sequence shown in table 6 or the hybridization of its complement (for example, with) with solid substrate." microarray " is to be minimized to require microexamination or by microexamination, to assist to estimate or the array of analysis.Array is called as DNA chip or biochip sometimes.

The array of molecule (" feature ") makes disposablely to be analyzed very in a large number and becomes possibility sample.In some example array, one or more molecules (for example oligonucleotide probe) will be repeatedly (for example twice) appear on array, for example so that internal contrast to be provided.

In particular instance, array comprises nucleic acid molecules, for example oligonucleotide sequence.The polynucleotide that use on array can be the common approximately cDNA of 500 to 5000 bases (" cDNA array ") of length, but also can use shorter or longer cDNA.Perhaps, polynucleotide can be the common approximately oligonucleotides of 20 to 80 bases of length, but shorter and longer oligonucleotides also is fit to.In an example, molecule comprises by its 5 ' end or 3 ' end and is attached to the oligonucleotides of array.

In array, each array sample is addressable, because its position can be stablized at least two dimensions of array and as one man determine.On array, the number of addressable point can change, for example from least 4 at least 9, at least 10, at least 14, at least 15, at least 20, at least 30, at least 50, at least 75, at least 100, at least 150, at least 200, at least 300, at least 500, at least 550, at least 600, at least 800, at least 1000, at least 10,000 or more.On array, different shapes can be taked in the feature application position.For example, array can be regular (for example with unified row and column, arranging) or irregular.Therefore, in orderly array, when sample is applied to array, distribute the position of each sample for sample, and can provide key (key) so that each position is related with suitable target or feature locations.Usually, oldered array is arranged with the lattice of symmetry, but sample can other patterns (for example with the line of radial distribution, helix or orderly bunch), arranges.Addressable array is normally computer-readable, the information about sample of the particular address on array and this position because computing machine can be stylized (for example hybridize or, in conjunction with data, comprise for example signal intensity) association.In some examples of computer-reader form, in array, personal feature is so that for example the Cartesian mesh model is regularly arranged, and it can be related with address information by computing machine.

In conjunction with or stable bond: the combination between two kinds of materials or molecule, the for example combination of a nucleic acid and another nucleic acid (for example combination of transcript shown in probe and table 6 or its complement), the perhaps combination of protein and another protein or nucleic acid molecules.Can detect combination by any program well known by persons skilled in the art, for example in the situation that nucleic acid for example passes through target: the physics of oligonucleotide complex or functional character.

The physical method that detects the combination of nucleic acid molecule complementation chain includes but not limited to for example following methods: DNA enzyme I or chemical footprinting, Gel migration and affinity cracking mensuration, Northern trace, Dot blot and light absorption trace routine.For example, a kind of method relates to observation and contains the solution of oligonucleotides (or analog) and target nucleic acid with the variation of the slow rising of temperature in the light absorption at 220 to 300nm places., if oligonucleotides or analog be in conjunction with its target, when oligonucleotides (or analog) and target dissociate each other or unwind, in the absorption of characteristic temperature, increase suddenly.In another example, described method relates to detection signal, the detectable label that for example exists on one or two nucleic acid molecules (or suitably time, antibody or protein).

Temperature (the T when combination between oligomer and target nucleic acid thereof is unwind by 50% oligomer and its target usually _m) characterize.With respect to having lower (T _m) compound, higher (T _m) the stronger or more stable compound of expression.

CDNA (complementary DNA): lack inner non-coding section (introne) and determine the section of DNA of the adjusting sequence transcribe.CDNA can synthesize by reverse transcription from the mRNA (mRNA) of cell and/or tissue samples extraction, and described sample is the colon sample for example, comprises the colon cancer sample.

Clinical effectiveness: refer to the patient in the situation that the health status for the treatment of is treated afterwards or do not had to disease or illness.Clinical effectiveness include but not limited to dead before the life-span increase, dead before the life-span reduce, chance of surviving increases, mortality risk increases, survival, without disease survival, chronic disease, transfer, late period or aggressive disease, palindromia, death with to the response of the favourable of therapy or difference.

Colon cancer: the cancer that forms in colon's (the longest part of large intestine).Most of colon cancers are gland cancer (starting and have the cancer of body of gland sample character in the cell that forms the internal).Cancer progression is characterized by cancer degree in phase or health.Usually whether contain cancer based on tumor size, lymph node and whether cancer diffuses to other parts of health from original position by stages.The phase of colon cancer comprises I phase, II phase, III phase and IV phase.Except as otherwise noted, the term colon cancer refers to the colon cancer of 0 phase that was in, I phase, II phase (comprising IIA or IIB phase), III phase (comprising IIIA, IIIB or IIIC phase) or IV phase.In some embodiments of this paper, colon cancer is from any phase.In other embodiments, colon cancer is II phase colon cancer.

Chemotherapeutics: treatment validity is arranged in the disease treatment that is characterized as abnormal cell growth any chemical reagent.The disease that such disease comprises tumour, neoplasm and cancer and is characterized as the hyperplasia growth is psoriasis for example.In one embodiment, chemotherapeutics is the medicament that is used for the treatment of colon cancer.In one embodiment, chemotherapeutics is the radioactivity compound.Those skilled in the art can easily identify useful chemotherapeutics (referring to for example, Slapak and Kufe, Principles of Cancer Therapy, Harrison's Principles of Internal Medicine, the 86th chapter, the 14th edition; The people such as Perry, Chemotherapy, Abeloff, Clinical Oncology the 2nd edition, the 17th chapter, 2000Churchill Livingstone, Inc; Baltzer and Berkery. (editor): Oncology Pocket Guide to Chemotherapy, the 2nd edition .St.Louis, Mosby-Year Book, 1995; Fischer Knobf, and Durivage (editor): The Cancer Chemotherapy Handbook, the 4th edition .St.Louis, Mosby-Year Book, 1993).The chemotherapeutics that is used for the treatment of colon cancer comprises little molecule for example 5 FU 5 fluorouracil, formyl tetrahydrofolic acid (leuvocorin), Irinotecan (irinotecan), oxaliplatin (oxaliplatin) and capecitabine (capecitabine), and antibody for example bevacizumab (bevacuzimab) and Cetuximab (cetuximab).Combination chemotherapy is to use to surpass a kind of medicament with the treatment cancer.

Contact: make and be in direct physical union; Comprise solid and liquid form.Contact comprises contacting between a molecule and another molecule, for example; Sample is contacted, any one probe of nucleic acid probe example sequence as shown in table 6 with nucleic acid probe.

Contrast: " contrast " refers to for sample or standard with experiment sample comparison, and described experiment sample is for example available from the tumor sample of colorectal cancer patients.In some embodiments, contrast is the sample available from healthy patients, perhaps available from the patient's who is diagnosed with colon cancer non-cancer tissue sample, for example from the non-cancer tissue sample of the homolog that wherein has tumour (for example, non-cancer colon can be used as the contrast of colon cancer).In some embodiments, contrast is historical control or standard value (that is, the check sample of test or represent baseline or the sample group of normal value) before.

Be used for comprising the sample that is considered to normal (because their desired character does not change, for example from the experimenter's who does not have colon cancer sample) and experiment value with the contrast of determining differential expression or standard with the sample comparison, even may be any group.Experimental standard and value can be based on population values known or that measure, and can allow figure or the tableau format of the measured value of experiment that compares and measures to provide.

Detect and express: measure horizontal expression with qualitative or quantitative manner and can detect nucleic acid.Exemplary method comprises microarray analysis, RT-PCR and Northern trace.In some instances, detect to express one or more the expression that comprises detection table 6 transcription thing.

The expression of differential expression or change: gene is (for example from any one of table 1,2 gene, and/or the transcribed nucleic acid thing in table 6) in, the information of coding changes into mRNA, mRNA changes into protein or the difference in both, for example increases or reduces.In some instances, difference is with respect to contrast or reference value, for example, from the expression of organizing the amplifying nucleic acid transcript that is not subjected to disease (for example colon cancer) impact of same subject, perhaps there is no the amount of expecting in the different experimenters of colon cancer.Difference can also, at the non-cancer tissue from experimenter's (having cancer in homolog), be compared from the different experimenters' that do not suffer from colon cancer tissue.Checkout discrepancy is expressed and can be comprised measuring in gene or protein expression and change, for example variation of the expression of one or more transcripts shown in one or more expression of listed gene and/or table 6 in table 1,2.

Lower or reduce: while when the reference nucleic acid molecule expression, using, guidance causes nucleic acid and generates any process that reduces.Gene outcome can be RNA (for example mRNA, rRNA, tRNA and structure RNA) or protein.Therefore, gene downward or inactivation comprise the process that reduces genetic transcription or mRNA translation.

Gene is lowered any detectable minimizing that comprises that gene outcome generates.In some instances, the generation of gene outcome and contrast (for example gene expression amount, for example gene expression of normal cell Plays) compared and is reduced by at least 1.2 times, for example at least 2 times, at least 3 times or at least 4 times.In several examples, contrast is the relative quantity that there is no gene expression in one or more experimenters of colon cancer or protein expression, for example without any the relative quantity of gene expression or protein expression in " without the cancer " experimenter of known cancer.

Extron: in theory, the section of the interruption gene that shows in the mRNA product.In theory, term " introne " but refer to be transcribed by the extron montage with its either side together and any DNA section from removing in transcript.In operation, exon sequence appears in the mRNA sequence of gene of No. ID, reference sequences (Ref.Seq) definition.In operation, intron sequences is the interior intervening sequence of genomic DNA of gene, by exon sequence, is comprised and at its 5' and 3' border, is had GT and AG montage consensus sequence.

Express: the coded message of gene is converted into the process of operation, not operation or the structure division of cell, and for example nucleic acid or protein is synthetic.Gene expression can be subjected to the impact of external signal.For example, cell is exposed to hormone can stimulate the expression of the gene of hormone induction.Dissimilar cell can the identical signal of difference ground response.Can also regulatory gene expression Anywhere the path from DNA to RNA to protein.Adjusting can comprise transcribing, the control of translation, rna transport and processing, middle element for example mRNA degraded or by activation, inactivation, compartmentation or the degraded of specified protein molecule after generating.

The expression of nucleic acid molecules can change, for example with respect to the expression in normal (for example, non-cancer) sample.The change of gene expression, for example differential expression, include but not limited to: the expression of (1) mistake; (2) the low expression; Or (3) expression inhibiting.The change that nucleic acid molecules is expressed can be followed and cause that in fact the expression of respective egg white matter changes." expression " and/or " relative expression " can be considered to specific transcript with respect to the expression value after threshold criteriaization, and described threshold value defines in the expression situation of every other transcript during for example colon cancer is expressed signature is signed in expression.Use the overall expression data of the given sample of method known to those skilled in the art standardization, with the extraction of the parent material for the difference amount, different efficiency and amplified reaction etc., proofread and correct.Use is diagnosed for the linear classifier of standardized data or prognosis (is for example conversed, good or poor prognosis) in fact mean the partition data space, that is, will sign by the property separated lineoid in all possible combination of expression value of all genes be divided into two half parts of separating.This is segmented on experience the training example that comes from large group, for example from the patient with good and poor prognosis.There is no general loss, can be the value of a certain fixedly group of all gene supposition except, this will define threshold value automatically for this remains gene, wherein determine and will change, for example from good prognosis to poor prognosis.Prognosis (for the gene with negative weight) or poor prognosis (for the gene with positive weight) have been indicated higher than the expression value of this dynamic threshold.The exact value of this threshold value depends on the express spectra of the actual measurement of every other gene in signature, it is fixing that but the general indication of some gene keeps, that is, high value or " the relative mistake expressed " always promote poor prognosis to determine that (gene with positive weight) or good prognosis determine (gene with negative weight).Therefore, in the situation of overall gene expression signature, whether the relative expression can indicate the rise of a certain transcript or lower and indicate or poor prognosis.

Gene magnification: form the gene of a plurality of copies or the process of genetic fragment in specific cells or clone.Replication region (DNA of one section amplification) is commonly called " amplicon ".Usually, the amount of the mRNA of generation (mRNA), that is, and the level of gene expression, the proportional increase of copy number that also forms with the specific gene of expressing.

Express spectra (or fingerprint or signature): gene expression pattern, it is the feature of specified disease stage or specific prognosis result or associated.Gene expression signature can represent by one group of information gene or its transcript, coding or non-coding or both.Can be used but not limited to method provided herein and estimate the expression of transcript in signature to make prognosis mensuration.Gene expression dose can be used for distinguishing for two kinds of clinical states of diagnosis or result normal and illing tissue for example, or for response and the non-response of method of prognosis, and about recurrence and the non-recurrence of Forecasting Methodology.But the gene expression (for example cDNA or mRNA) that the gene expression of difference or change can be by detection limit but variation or detected by the variation of the protein of the detection limit of those gene expressions.Different or appraisable gene expression pattern, for example determine the gene of organizing or gene indication nucleic acid for example height and the low expression pattern of EST; In some instances, few as one or two gene, provide spectrum, but can use more polygenes in composing, such as at least 2, at least 3, at least 4, at least 5, at least 6, at least 7, at least 9, at least 10 or at least 11 etc.In some embodiments, described spectrum comprises at least about 200 genes (or " transcript ") and about 1000 transcripts at the most, for example from about 400 transcripts to about 800 transcripts, or approximately 500 transcripts to about 700 transcripts.Described spectrum comprises the transcript (for example, from least 100, at least 200, at least 300, at least 400, at least 500 or at least 600 transcripts of table 6) from table 6, comprises in some embodiments 636 transcripts that table 6 is listed.As used herein, term " gene " refers to the transcript of expressing, and it can be the gene that characterizes, and can be perhaps the transcript of expressing, for example EST.In some embodiments, detection platform is microarray, and each probe is considered to measure the expression of " gene " or " transcript " separately.

Gene expression profile (also being called fingerprint or signature) can related tissue or the moment of cell type (for example colon), normal structure growth or progression of disease (for example colon cancer) or affect any other different or identifiable state of gene expression in measurable mode.Gene expression profile can comprise relative expression and the absolute expression levels of specific gene, and can consider in the situation of the test sample book of with baseline or check sample spectrum (for example from the experimenter who does not suffer from colon cancer sample), comparing.In an example, experimenter's gene expression profile is read to state on array (for example nucleic acid array).

Hybridization: form base-pair between the complementation district of two DNA chains, RNA chain or DNA chain and RNA chain, thus formation duplex molecule, the duplex that for example forms between any one or its complement of nucleotide sequence shown in probe and table 6.Cause the hybridization conditions of specific strict degree to change according to the character of hybridizing method and composition and the length of hybrid nucleic acid sequence.Generally speaking, the ionic strength of hybridization temperature and hybridization buffer (Na for example ⁺Concentration) will determine the hybridization severity.People such as Sambrook, (1989) Molecular Cloning, second edition, Cold Spring Harbor Laboratory, Plainview, discussed the calculating of the hybridization conditions that reaches specific strict degree in NY (the 9th and 11 chapter).Be below exemplary hybridization conditions group, and be not restrictive:

Very high severity (sequence of at least 90% homogeneity is enjoyed in detection)

Hybridization: 5x SSC continues 16 hours under 65 ℃

Washed twice: 2x SSC continues 15 minutes at every turn under room temperature (RT)

Washed twice: 0.5x SSC continues 20 minutes at every turn under 65 ℃

High severity (sequence of at least 80% homogeneity is enjoyed in detection)

Hybridization: 5x-6x SSC continues 16-20 hour under 65 ℃-70 ℃

Washed twice: 2x SSC continues 5-20 minute at every turn under RT

Washed twice: 1x SSC continues 30 minutes at every turn under 55 ℃-70 ℃

Low severity( The sequence of at least 60% homogeneity is enjoyed in detection)

Hybridization: 6x SSC continues 16-20 hour under RT to 55 ℃

Washing at least twice: 2x-3x SSC continues 20-30 minute at every turn under RT to 55 ℃

Separate: " separation " biological components (for example nucleic acid molecules, protein or cell) with the natural existence of described component biological cell or the isolated or purified in fact of the other biological component in biosome itself wherein, described other biological component is other chromosomes and extrachromosomal DNA and RNA, protein and cell for example.This term also comprises in host cell by the nucleic acid molecules of recombinant expressed preparation and the nucleic acid molecules of chemosynthesis.For example, the cell of separation, for example colon cancer cell, be the cell that separates in fact with the cell of other types.

Label: can for example pass through the material that ELISA, spectrophotometry, flow cytometry or microscope or other visualization techniques detect.For example, label can be connected to nucleic acid molecules or protein, thereby allows to detect nucleic acid molecules or protein.For example, for example nucleic acid molecules or the antibody of target nucleic acid molecule of specific binding target molecule.The example of label includes but not limited to radioactive isotope, zymolyte, co-factor, part, chemiluminescence agent, fluorophore, haptens, enzyme and combination thereof.Labeling method and about the guidance of the label of selecting to be fit to various purposes, be discussed at (the Molecular Cloning:A Laboratory Manual such as people such as Sambrook, Cold Spring Harbor, New York, 1989) and people (Current Protocols in Molecular Biology, the John Wiley﹠amp such as Ausubel; Sons, New York, 1998).

Long-term surviving: for the operation of colon cancer or other treatment (for example, chemotherapy) afterwards, without disease survival at least 3 years, more preferably at least 5 years, even more preferably at least 8 years.

More aggressive: as used herein, the colon cancer of " more aggressive " form is to have the transfer of relative increase or the colon cancer of risk of recurrence (for example after the exenterate tumour)." more aggressive " colon cancer can also refer to give the colon cancer in life-span before dead possibility that the colon cancer experimenter increases or minimizing dead.Suffering from more aggressive " experimenter of the colon cancer of form is considered to high risk (poor prognosis).

Represent the nucleic acid molecules of gene: any length and any nucleic acid that provide corresponding gene information that is suitable as probe or other indication molecules, for example DNA (introne or extron or both), cDNA or RNA (for example mRNA), those that for example list in table 1 or 2, the transcript of for example listing in table 6.

Oligonucleotides: relatively short polynucleotide include but not limited to strand deoxyribonucleotide, strand or double-stranded ribonucleotide, RNA:DNA crossbred and double-stranded DNA.Often pass through chemical method, but the automated oligonucleotide synthesizer that for example uses commercial sources to obtain, synthetic oligonucleotide, for example ssDNA probe oligonucleotides.Yet, can prepare oligonucleotides by many additive methods, comprise the technology of extracorporeal recombinant DNA mediation and by express DNA in the Cell and organism body.

The patient: as used herein, term " patient " comprises people and non-human animal.Preferred patient for treatment is the people." patient " and " experimenter " is used interchangeably at this paper.

Patient's response: can assess any terminal of patient's benefit with indication, include but not limited to that (1) suppresses tumor growth to a certain extent, comprise and slowing down and growth retardation fully; (2) the tumour cell number reduces; (3) tumor size reduces; (4) inhibition (that is, reduce, slow down or stop fully) tumour cell is invaded profit and is entered adjacent peripheral organ and/or tissue; (5) suppressing (that is, reduce, slow down or stop fully) shifts; (6) strengthen anti-tumor immune response, its can but not necessarily lead and oncogenicly disappear or suppress; (7) alleviate to a certain extent one or more symptoms with Tumor-assaciated; (8) after the treatment, the survival duration increases; And/or (9) mortality ratio that preset time, point reduced after treatment.

Polynucleotide: when with odd number or plural number, using, refer generally to any polyribonucleotide or polydeoxyribonucleotide, it can be the RNA of unmodified or RNA or the DNA of DNA or modification, perhaps even its combination.Therefore, for example, polynucleotide defined herein include but not limited to strand and double-stranded DNA, the DNA that comprises strand and double-stranded region, strand and double-stranded RNA,, with the RNA that comprises strand and double-stranded region, comprise it can being strand or more generally double-stranded or comprise the DNA of strand and double-stranded region and the hybrid molecule of RNA.Term " polynucleotide " also comprises DNA and the RNA that contains one or more modified bases.Therefore, DNA or RNA with skeleton of modifying for stability or other reasons are " polynucleotide ", and be as desired at this paper in this term.And the DNA or the RNA that comprise uncommon base (for example inosine) or modified base (for example tritiate base) are included in term defined herein " polynucleotide ".Generally speaking, term " polynucleotide " comprises the form that all chemistry, enzyme and/or the metabolism of the polynucleotide of unmodified are modified, and virus and the DNA of cell and the chemical species of RNA feature, and described cell comprises simple and complex cell.

Probe and primer: probe comprises the nucleic acid that separates that can hybridize with target nucleic acid (one of routine nucleotide sequence as shown in table 6 or its complement).Detectable mark or reporter molecules can be connected to probe.Typical mark comprises radioactive isotope, zymolyte, co-factor, part, chemiluminescence and or fluorescer, haptens and enzyme.The method of preparation and use nucleic acid probe and primer is described in (the Molecular Cloning:A Laboratory Manual such as people such as Sambrook, CSHL, New York, 1989), the people such as Ausubel (editor) (Current Protocols in Molecular Biology, John Wiley﹠amp; Sons, New York, 1998), and the people (PCR Protocols, A Guide to Methods and Applications, Academic Press, Inc., San Diego, CA, 1990) such as Innis.The selection of the mark of labeling method and suitable various purposes is instructed and is discussed at (the Molecular Cloning:A Laboratory Manual such as people such as Sambrook, CSHL, New York, 1989) and people (Current Protocols in Molecular Biology, the John Wiley﹠amp such as Ausubel; Sons, New York, 1998).

probe length is generally at least 12 nucleotide, for example at least 12, at least 13, at least 14, at least 15, at least 16, at least 17, at least 18, at least 19, at least 20, at least 21, at least 22, at least 23, at least 24, at least 25, at least 26, at least 27, at least 28, at least 29, at least 30, at least 31, at least 32, at least 33, at least 34, at least 35, at least 36, at least 37, at least 38, at least 39, at least 40, at least 45, at least 50 or continuous nucleotide more and the target nucleic acid molecule complementation, described target nucleic acid molecule is 15-50 nucleotide for example, the primer of 20-50 nucleotide or 15-30 nucleotide.In some instances, probe is even longer, cDNA probe for example, and its length can be from approximately 500 to surpassing 5000 nucleotide.

Primer is short nucleic acid molecules, and for example length is 10 oligonucleotides or more DNA oligonucleotides, and it can be annealed to form crossbred between primer and target nucleic acid chain by the target nucleic acid molecule of nucleic acid hybridization and complementation.Primer can extend along target nucleic acid molecule by polymerase.Therefore, primer can be used for amplifying target nucleic acid molecule (nucleotide sequence that example is as shown in table 6).

The specificity of primer and/or probe increases with its length.Therefore, for example, comprise specificity and target sequence annealing that the primer of 30 continuous nucleotides will be higher with the corresponding primer than 15 nucleotide only.Therefore,, in order to obtain larger specificity, can select to comprise at least 15,20,25,30,35,40,45,50 or probe and the primer of more continuous nucleotides.In particular instance, primer length is at least 15 nucleotide, for example with at least 15 continuous nucleotides of target nucleic acid molecule complementation.the length-specific that can be used for implementing the primer of disclosure method comprises with target nucleic acid molecule complementation to be amplified, have at least 15, at least 16, at least 17, at least 18, at least 19, at least 20, at least 21, at least 22, at least 23, at least 24, at least 25, at least 26, at least 27, at least 28, at least 29, at least 30, at least 31, at least 32, at least 33, at least 34, at least 35, at least 36, at least 37, at least 38, at least 39, at least 40, at least 45, at least 50 or the primer of more continuous nucleotides, 15-50 nucleotide for example, the primer of 20-50 nucleotide or 15-30 nucleotide.One of greatest factor of considering in the PCR design of primers comprises primer length, melting temperature (Tm) and GC content, specificity, complementary primer sequence and 3' terminal sequence.Generally speaking, best PCR primer length is generally 17-30 base, and contains the 20-80% that has an appointment, for example about 50-60%G+C base.Between 50 ℃ and 80 ℃, for example approximately the Tm of 50 ℃ to 70 ℃ is normally preferred.

Primer pair can be used for amplifying nucleic acid sequence, for example by PCR, PCR in real time or other nucleic acid amplification methods known in the art." upstream " or " forward " primer is the primer of reference point 5' end on nucleotide sequence." downstream " or " oppositely " primer is the primer of reference point 3' end on nucleotide sequence.Generally speaking, at least one forward and a reverse primer are included in amplified reaction.

Nucleic acid probe and primer can easily prepare based on nucleic acid molecules provided herein, for example, by use, expect the computer program that is used for this purpose, for example Primer (version 0.5,

1991, Whitehead Institute for Biomedical Research, Cambridge, MA) or PRIMER

Software (Applied Biosystems, AB, Foster City, CA).

Further guidance for PCR primer and probe design can be referring to people such as Dieffenbach, General Concepts for PCR Primer Design, in: PCR Primer, A Laboratory Manual, Cold Spring Harbor Laboratory Press, New York, 1995,133155 pages; Innis and Gelfand, Optimization of PCRs, in: PCR Protocols, A Guide to Methods and Applications, CRC Press, London, 1994,511 pages; And Plasterer, Primerselect:Primer and probe design.Methods Mol.Biol.70:520 527,1997.

Prognosis: the possibility of clinical effectiveness of suffering from the experimenter of specified disease or illness.With regard to cancer, prognosis represents experimenter the survive possibility (probability) of (for example 1,2,3,4 or 5 year) and/or the possibility (probability) of metastases.Term " prediction " is used to refer to the patient advantageously or adversely responds a kind of medicine or the possibility of one group of medicine and also refer to the degree of those responses at this paper.By being that any particular patient is selected optimal therapeutic modality, Forecasting Methodology of the present invention can be used for clinically making treatment and determine.Advantageously respond therapeutic scheme at prediction patient possibility, for example operation get involved, with chemotherapy and/or the radiotherapy aspect of given medicine or drug regimen, Forecasting Methodology of the present disclosure is valuable instrument.

Purifying: term " purifying " does not require absolute purity; But it is expected as relative terms.Therefore, for example, the oligonucleotides goods of purifying be wherein oligonucleotides than purer oligonucleotides goods under the environment that comprises complicated oligonucleotide mixture.

Sample: available from experimenter's the biological sample that contains genomic DNA, RNA (comprising mRNA and Microrna), protein or its combination.Example includes but not limited to peripheral blood, urine, saliva, biopsy, extract, specimens from pri and autopsy, and comprises fixing and/or paraffin-embedded sample.In an example, sample comprises the sample of colon biopsy (for example colon cancer tumour), non-cancer tissue or the sample of normal structure (from the experimenter who does not suffer from known disease or illness, for example without the cancer experimenter).

Sequence homogeneity/similarity: according to homogeneity between sequence or similarity, express homogeneity/similarity between two or more nucleotide sequences or two or more amino acid sequences.Can measure sequence homogeneity according to number percent homogeneity; Number percent is higher, and sequence is more same.Can measure sequence similarity according to number percent similarity (considering that conserved amino acid replaces); Number percent is higher, and sequence is more similar.

The comparison method that is used for sequence relatively is well known in the art.Various programs and alignment algorithm are described in: Smith﹠amp; Waterman, Adv.Appl.Math.2:482,1981; Needleman﹠amp; Wunsch, J.Mol.Biol.48:443,1970; Pearson﹠amp; Lipman, Proc.Natl.Acad.Sci.USA85:2444,1988; Higgins﹠amp; Sharp, Gene, 73:237-44,1988; Higgins﹠amp; Sharp, CABIOS5:151-3,1989; The people such as Corpet, Nuc.Acids Res.16:10881-90,1988; The people Computer Appls.the Biosciences8 such as Huang, 155-65,1992; With the people such as Pearson, Meth.Mol.Bio.24:307-31,1994.The people such as Altschul, J.Mol.Biol.215:403-10,1990, the detailed consideration that provides sequence alignment method and homology to calculate.

The basic Local Alignment gopher of NCBI (the BLAST) (people such as Altschul, J.Mol.Biol.215:403-10,1990) can be available from several sources, comprise national biotechnology center (NCBI, National Library of Medicine, Building38A, Room8N805, Bethesda, MD20894) and internet on, be combined with sequential analysis program blastp, blastn, blastx, tblastn and tblastx.Other information can find on the NCBI website.

BLASTN is used for relatively nucleotide sequence, and BLASTP is used for the comparing amino acid sequence.If the sequence of two comparisons is enjoyed homology, the output file of appointment will propose those zones with the aligned sequences homology.If the sequence of two comparisons is not enjoyed homology, the output file of appointment will not propose the sequence of comparing.

Once comparison, by counting wherein identical nucleotide or the amino acid residue number that comes across two positions in sequence determine to mate number.By the sequence length that provides in the sequence of coupling number divided by evaluation, perhaps divided by clear and definite length (for example, from 100 continuous nucleotides identifying the sequence that provides in sequence or amino acid residue), what then obtain is on duty with 100, determines number percent sequence homogeneity.For example, when the cycle tests with having 1554 nucleotide was compared, nucleotide sequence and described cycle tests with 1166 couplings were 75.0% same (1166 ÷ 1554*100=75.0).Number percent sequence homogeneity value is rounded to tenths.For example, 75.11,75.12,75.13 and 75.14 are given up to 75.1, and 75.15,75.16,75.17,75.18 and 75.19 quilts are entered to 75.2.Length value is integer always.In another example, the target sequence that contains the zone of 20 nucleotide comparing with 20 continuous nucleotides of sequence from following evaluation contains the zone (that is, 15 ÷ 20*100=75) of enjoying 75% sequence homogeneity with this evaluation sequence.

Two closely-related indications of nucleic acid molecules are the phase mutual crosses under stringent condition as above of two molecules.Yet, due to the degeneracy of genetic code, may the encode amino acid sequence of same or similar (conservative) of the nucleotide sequence that does not show height homogeneity.Can use this degeneracy to change in nucleotide sequence, to produce a plurality of nucleic acid molecules of the essentially identical protein of whole codings.For example, by the method, measure, this homologous nucleotide sequence can have at least about 60%, 70%, 80%, 90%, 95%, 98% or 99% sequence homogeneity with the listed molecule of table 6.

It will be understood by those skilled in the art that specific sequence homogeneity scope only provides for guidance; May obtain to fall into the very significantly homologue outside the scope that provides.

Montage or RNA montage: remove introne and connect extron and enter with generation that eukaryotic is cytoplasmic, the RNA processing of ripe mRNA with continuous coded sequence.

Transcript or gene outcome: by the process of from its corresponding DNA or cDNA template, transcribing, produce or derivative RNA molecule.Transcript comprises coding and non-coding RNA molecule, such as but not limited to mRNA, rRNA (rRNA), the transfer RNA (tRNA) of mRNA (mRNA), alternative splicing and large-scale other transcripts that are not translated into protein, small nuclear rna (snRNA), antisense molecule other rna transcription things of short interfering rna (siRNA) and Microrna (miRNA) and unknown function for example for example.In some embodiments, transcript is the nucleotide sequence shown in table 6.

Treatment: the generic term that comprises diagnosis and treatment.

Treatment: comprise therapeutic treatment and preventative or preventive measure, wherein purpose is prevent or slow down (alleviating) target pathological state or illness.Those that need treatment comprise suffer from illness those and be easy to suffer from illness those maybe to prevent to suffer from those of illness.In tumour (for example cancer) treatment, treat for example operation, chemotherapy or radiation and can directly reduce the pathology of tumour cell, perhaps make tumour cell more responsive to further treatment.

Tumour, neoplasia, malignant tumour or cancer: no matter the neoplastic cell Growth and reproduction is pernicious or optimum, and the result of the abnormal and uncontrolled growth of all front cancers and cancer cell and tissue and cell.Term " cancer " and " cancer " refer to or describe the mammiferous pathological state that usually by not modulated Growth of Cells, is characterized.Neoplasia, malignant tumour, cancer and tumour are often used interchangeably, and refer to tissue that excessive cell division causes or the misgrowth of cell.In individuality, the amount of tumour is the number that can be measured as tumour, " tumor load " of volume or weight.The tumour that does not have to shift is called as " optimum ".The tumour of invading surrounding tissue and/or can shift is called as " pernicious ".But " non-cancer tissue " is to form from malignant growth wherein there is no the tissue of the homolog of excrescent feature pathology.Generally speaking, non-cancer tissue is acted normally on histology." normal structure " is the tissue from organ, and wherein said organ is not affected by another disease or the illness of cancer or this organ." without cancer " experimenter also is not diagnosed with the cancer of this organ, and there is no detectable cancer.

" pathology " of cancer comprises all phenomenons of damaging patient health.Before this normal function, cell factor or other secretory product that includes but not limited to abnormal or uncontrolled Growth of Cells, transfer, interference flanking cell discharges, suppresses or increase the weight of inflammatory or immune response, neoplasia, malignant tumour with abnormal level, around malignant tumour, intrusion or at a distance tissue or organ, such as lymph node etc.

Tumour-knot-transfer (TNM): the TNM classification of malignant tumour is for the cancer staging system of describing patient body cancer degree.T describes the big or small of primary tumor and whether it has invaded near tissue; N describes any lymph node that relates to; Describe and shift with M.TNM is developed by International Union Against Cancer (International Union Against Cancer) and safeguards to realize the common recognition to global the recognized standard of classification cancer dispersion level.

Raise or activate: when reference nucleic acid molecule is expressed use, instruct any process that gene outcome generates to be increased that causes.Gene outcome can be RNA (for example mRNA, rRNA, tRNA and structure RNA) or protein.Therefore, gene raises or activates the process that comprises increase genetic transcription or mRNA translation, for example proinflammatory gene.

The example that increases the process transcribe comprises those that promote that the transcription initiation compound forms, increase transcription initiation speed those, increase transcribe those that extend speed, increase transcribe synthesis capability those and alleviate those of transcribing inhibitions (for example, by blocking the combination of transcription repression).Gene raises and can comprise the inhibition of checking and stimulate expression higher than existing level.Those that the example that increases the process of translation comprises those that increase translation initiation, increase that translation extends and increase those of mRNA stability.

Gene raises any detectable increase of the generation that comprises gene outcome, for example proinflammatory gene.In some instances, at least 1.2 times of increases, for example at least 2 times, at least 3 times, at least 4 times, at least 5 times, at least 8 times, at least 10 times or at least 15 times are compared in the generation of gene outcome with contrast (for example amount of gene expression and/or standardization gene expression in normal cell).

Weight:, with reference to gene signature disclosed herein, refer to the relative importance of a project in statistical computation, for example the relative importance of table 6 transcription thing.The weight of each transcript in gene expression signature can be determined according to clinical samples data group with analytical approach known in the art.Exemplary process is described hereinafter.

Be used for to implement or test suitable method of the present disclosure and material is described hereinafter.This method and material are only exemplary, and are not restrictive.Can use and those similar or be equal to additive method and materials described herein.For example, the known conventional method in the described field of the disclosure is described in various general and more specifically in list of references, comprises such as people such as Sambrook, Molecular Cloning:A Laboratory Manual, the 2nd edition, Cold Spring Harbor Laboratory Press, 1989; The people such as Sambrook, Molecular Cloning:A Laboratory Manual, the 3rd edition, Cold Spring Harbor Press, 2001; The people such as Ausubel, Current Protocols in Molecular Biology, Greene Publishing Associates, 1992 (with supplementary issue to 2000); The people such as Ausubel, Short Protocols in Molecular Biology:A Compendium of Methods from Current Protocols in Molecular Biology, the 4th edition, Wiley﹠amp; Sons, 1999; Harlow and Lane, Antibodies:A Laboratory Manual, Cold Spring Harbor Laboratory Press, 1990; Harlow and Lane, Using Antibodies:A Laboratory Manual, Cold Spring Harbor Laboratory Press, 1999Oligonucleotides Synthesis, (M.J.Gait edits, 1984); Animal Cell Culture, Freshney edits, and 1987; Methods in Enzymology, Academic Press, Inc.); Handbook of Experimental Immunology, the 4th [Dan, D.M.Weir﹠amp; C.C.Blackwell edits, Blackwell Science Inc., 1987; Gene Transfer Vectors for Mammalian Cells, J.M.Miller﹠amp; M.P.Calos edits, and 1987); With PCR:The Polymerase Chain Reaction, the people such as Mullis edit, and 1994.In addition, material, method and example are only exemplary, and are not restrictive.

II. the description of several embodiments

A. colon cancer is expressed signature and using method

Herein disclosed is the expression signature from colon cancer.Disclosed signature can be used for the prognosis of colon cancer, the diagnosis of colon cancer and the application in classification patient group.In some embodiments, available from the experimenter for example patient's sample be processed to one group of polynucleotide in conjunction with target, it represents the transcript of expressing in tissue samples.Survey polynucleotide in conjunction with target, to obtain the information of relevant transcript expression with representative or corresponding to the complementary polynucleotide probe of signature described herein.Randomly calculate the decision-making scoring, the expression of described decision-making scoring by procuration transcription thing.Then with decision-making scoring and contrast, for example patient's faciation compares, and sample similar in heredity is associated with known patient's response or clinical effectiveness.For example, also provide response and the prognosis after to treatment of colon cancer (for example excision and/or chemotherapy) with the prediction patient of sensitive method.Generally speaking, analysis of history patient group data and tissue samples have the patient's of the passing history of colon cancer heredity spectrum with generation.In some embodiments, convert the heredity of clinical samples spectrum to the decision-making scoring.Each patient's clinical effectiveness is composed with the heredity of each individual patients cancer or from the decision-making that this heredity spectrum mathematics is derived, is marked related.

In some embodiments, use known historic patient data to produce mathematical algorithm and be applied to Forecasting Methodology for the new patient who suffers from colon cancer.In some embodiments, according to choice criteria, such as patient's result, to the response of therapy and recurrence etc., algorithm produces the patient is divided into the threshold value of two groups.In some instances, before being used for Forecasting Methodology described herein, with further historic patient group data, verify mathematical algorithm or threshold value.Then, mathematical algorithm or threshold value can be used as reference, for example in contrast, and with the decision-making scoring of relatively from patient's heredity spectrum of thirsting for the colon cancer Forecasting Methodology, deriving.In some embodiments, these results allow to estimate separately or with the genome evidence of the effect of the operative treatment colon cancer of NACT combination.

Signature described herein can be effectively and can distinguish two diagnosis or prognosis result.An importance of the present disclosure is with some gene is measured in colon cancer tissue expression, patient and optimal treatment to be mated, and prognosis information is provided.

In some embodiments, use colorectal cancer to focus on microarray research tool exploitation signature.In a specific embodiment, this research tool is by Almac Diagnostics, and the colorectal cancer of expression data accurately that can transmit of Ltd. (Almac Diagnostics, Ltd., N.Ireland) exploitation is transcribed group focusing research array.

Colorectal Cancer DSA ^TMResearch tool contains 61,528 probe groups, and coding is proved to be 52,306 transcripts at colon cancer and normal tissue expression.Use BLAST to analyze, for the mankind of U.S. biotechnology information center (NCBI) reference sequences (RefSeq) RNA database (can available from WWW ncbi.nlm.nih.gov/RefSeq/), compared Colorectal Cancer DSA ^TMResearch tool, 21,968 (42%) individual transcripts are present in mankind RefSeq database, and 26,676 (51%) individual transcripts are not present in mankind RefSeq database.And 7% content represents the antisense transcript of the expression of note gene.(people such as Johnston, J.Clin.Oncol.24:3519,2006; The people such as Pruitt, Nucleic Acids Research33:D501-D504,2005).In addition, the Colorectal Cancer DSA that compares with main general array ^TMThe analysis of probe level, outstanding about 20,000 (40%) individual transcripts are not included on main general microarray platform (Affymetrix), and are Colorectal Cancer DSA ^TMUnique.Therefore, Colorectal Cancer DSA ^TMResearch tool comprises does not also have transcript available in the gene expression research of carrying out so far.

In some embodiments,, if expression increases or reduces between interested state, think that the expression of gene expression signature transcription thing has information.Can assess by method known to those skilled in the art increase or the minimizing of gene expression, described method includes but not limited to use a doubly change, t check, F check, Wilcoxon rank test, the ANOVA (people such as Cui, Genome Biology4:210,2003)) or for detection of the special method of differential expression, significance analysis (the people such as Tusher such as microarray, Proc.Natl.Acad.Sci.USA98:5116-21,2001)) or LIMMA (Smyth, Stat.Appl.Genet.Mol.Biol., 3:Art.3,2004)).

In some embodiments, the transcript in signature is used to form the weighting sum of its signal, and wherein individual weight can be positive or negative.That obtain and (" decisive function ") compares with predetermined reference point.With reference point relatively can be used for diagnosis or prediction clinical state or result.

The transcript that the signature that it will be appreciated by the skilled addressee that provides in table 1,2 and/or 6 comprises will carry the weight that does not wait in the signature of diagnosis or prognosis colon cancer.Therefore, although few as 1 sequence can be used for diagnosis or predicts the outcome, specificity and sensitivity or diagnosis or forecasting accuracy can increase with multisequencing more.The weight order classification transcript of table 6 to successively decrease in signature, be defined as the grade according to average weight in the compound decision-making score function of cross validation measurement.The weight grade is also corresponding to SEQ ID NO in appended sequence table:, the transcript that therefore has weight limit is SEQ ID NO:1.

in some embodiments, signature comprises at least 2 of table 6 transcription thing, for example at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 11, at least 12, at least 13, at least 14, at least 15, at least 20, at least 25, at least 30, at least 35, at least 40, at least 45, at least 50, at least 55, at least 60, at least 65, at least 70, at least 75, at least 80, at least 85, at least 90, at least 95, at least 100, at least 125, at least 150, at least 175, at least 200, at least 225, at least 250, at least 275, at least 300, at least 325, at least 350, at least 375, at least 400, at least 425, at least 450, at least 475, at least 500, at least 525, at least 550, at least 575, at least 600, at least 634 or even whole 636, it has the weight limit that is defined as the grade of average weight in the compound decision-making score function of measuring according to cross validation, and still has prognosis values.In some embodiments, signature comprises front 10 weighting transcripts that table 6 is listed, second front 10 weighting transcripts, the 3rd front 10 weighting transcripts, the 4th front 10 weighting transcripts, the 5th front 10 weighting transcripts, the 6th front 10 weighting transcripts, the 7th front 10 weighting transcripts, the 8th front 10 weighting transcripts, the 9th front 10 weighting transcripts or the tenth front 10 weighting transcripts.in other embodiments, signature comprises that table 6 is listed and has 636 of a weight limit, 634, 620, 610, 600, 590, 580, 570, 560, 550, 540, 530, 520, 510, 500, 490, 480, 470, 460, 450, 440, 430, 420, 410, 400, 390, 380, 370, 360, 350, 340, 330, 320, 310, 300, 290, 280, 270, 260, 250, 240, 230, 220, 210, 200, 190, 180, 170, 160, 150, 140, 130, 120, 110, 100, 90, 80, 70, 60, 50, 40, 30, 20 or 10 transcripts.In some embodiments, signature is based on approximately 200 to about 1000 transcripts, for example approximately 400 to about 800 transcripts, for example approximately 500 to about 700 transcripts or in some embodiments approximately 550 to the about expression of 650 transcripts, comprise as mentioned above from table 6 those (for example, from table 6 at least about 50, at least about 100, at least about 200, at least about 300, at least about 400, at least about 500 or at least about 600 or whole transcripts).

In one embodiment, specific signature can be used for method disclosed herein, and it comprises the transcript of MUM1 and SIGMAR1.In another embodiment, signature can be used for method disclosed herein, and it comprises the transcript of MUM1, SIGMAR1, ARSD, SULT1C2 and PPFIBP1.In another embodiment, signature can be used for method disclosed herein, and it comprises the transcript of ARSD, CXCL9, PCLO, SLC2A3, FCGBP, SLC2A14, SLC2A3, BCL9L and the antisense sequences of MUC3A, OLFM4 and RNF39.This signature is by following table 1 representative.

Show 10 candidate's core transcripts in 1:636 transcript label

In some embodiments, provide the genetic transcription thing of one group of core in the colon cancer signature, it is identified by independent research, to determine each contribution to the signature performance of 636 probe groups.In this embodiment, from 636 probe groups signatures, remove 10 probe groups, and use the training data group to produce new signature based on 636 probe groups.Then use new signature prediction verification msg group (without threshold value) and measure AUC.Record is from the difference of the AUC of 636 probe groups signatures.Repeat this process 500,000 times, and record causes lacking the mean difference of AUC of the signature of described probe groups.Recorded the probe groups with maximum negative Δ AUC in table 1.In this embodiment, 10 transcripts of this group represent the gene of candidate's core group, and its disappearance from signature is significantly damaged the estimated performance of signature.Therefore, in certain embodiments, represent that the transcript of the gene in table 1 is included in the colon cancer signature.In table 1, if this transcript saves from signature, the decline of DAUC representative checking AUC.Direction has been described the direction of the transcript of expressing in the colon.Three transcripts in this signature are represented as the antisense transcript of MUC3A, OLFM4 and RNF39.

In some embodiments, signature comprises that it comprises ARSD, CXCL9, PCLO, SLC2A3, FCGBP, SLC2A14, SLC2A3, BCL9L, MUC3A, OLFM4 and RNF39 from the combination of 626-636 transcript of table 6.And in another embodiment, signature comprises transcript 10-636,10-50,50-636, the 100-636 that table 6 is listed, it comprises ARSD, CXCL9, PCLO, SLC2A3, FCGBP, SLC2A14, SLC2A3, BCL9L, MUC3A, OLFM4 and RNF39, and wherein the transcript direction is indicated in table 6.

It should be noted that by main general array, by the probe horizontal analysis, with 176 transcripts be accredited as not representative (that is, they are above-mentioned Colorectal Cancer DSA ^TMInstrument institute " unique ").Listed 176 transcripts of this group of table 2 are described as transcript to this paper colon gene signature and using method uniqueness at this paper.Probe-sequence Ortholog retrieval has been accredited as these transcripts and has not been included in main general array (Affymetrix) upper (that is, they are above-mentioned Colorectal Cancer DSA ^TMResearch tool institute " unique ").Many these transcripts are not reported the antisense transcript of expressing before being.These 176 transcripts are provided in the following table 2, and wherein the weight grade is corresponding to the numbering shown in table 6.Therefore, the sequence of these unique transcripts can be referring to table 6.

Show the unique transcript in 2:636 transcript label

in some embodiments, signature comprises at least 2 of listed transcript in table 2, for example at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 11, at least 12, at least 13, at least 14, at least 15, at least 20, at least 25, at least 30, at least 35, at least 40, at least 45, at least 50, at least 55, at least 60, at least 65, at least 70, at least 75, at least 80, at least 85, at least 90, at least 95, at least 100, at least 125, at least 150 or even whole 176, for example have the weight limit that is defined as the grade of average weight in the compound decision-making score function of measuring according to cross validation and still have those of prognosis values.In some embodiments, signature comprises front 10 weighting transcripts that table 2 is listed, second front 10 weighting transcripts, the 3rd front 10 weighting transcripts, the 4th front 10 weighting transcripts, the 5th front 10 weighting transcripts, the 6th front 10 weighting transcripts, the 7th front 10 weighting transcripts, the 8th front 10 weighting transcripts, the 9th front 10 weighting transcripts or the tenth front 10 weighting transcripts.And in other embodiments, signature comprises listed 176,160,150,140,130,120,110,100,90,80,70,60,50,40,30,20 or 10 transcripts with weight limit of table 2.

In some embodiments, method described herein comprises making to separate and stands gene expression profile from patient's RNA.Therefore, can complete gene expression profile for one group of gene, described one group of gene comprises at least two in the listed transcript of table 6, and it is standardization as described below in some instances.in the particular of method disclosed herein, measured table 6 transcription thing at least 2, for example at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 11, at least 12, at least 13, at least 14, at least 15, at least 20, at least 25, at least 30, at least 35, at least 40, at least 45, at least 50, at least 55, at least 60, at least 65, at least 70, at least 75, at least 80, at least 85, at least 90, at least 95, at least 100, at least 125, at least 150, at least 175, at least 200, at least 225, at least 250, at least 275, at least 300, at least 325, at least 350, at least 375, at least 400, at least 425, at least 450, at least 474, at least 500, at least 525, at least 550, at least 575, at least 600, at least 634 or the expression of even whole 636 or its expression product and/or complement, for example, has the weight limit that is defined as average weight grade in the compound decision-making score function of measuring according to cross validation and still have transcript in the table 6 of prognosis values.in some embodiments of the method, measured table 2 transcription thing at least 2, for example at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 11, at least 12, at least 13, at least 14, at least 15, at least 20, at least 25, at least 30, at least 35, at least 40, at least 45, at least 50, at least 55, at least 60, at least 65, at least 70, at least 75, at least 80, at least 85, at least 90, at least 95, at least 100, at least 125, at least 150 or the expression of even whole 176 or its expression product and/or complement, for example, have the weight limit that is defined as average weight grade in the compound decision-making score function of measuring according to cross validation and still have those of prognosis values.In method described herein, the combination of transcript can be called as signature or express signature.

Measure relative expression's level of colon's transcription thing to form gene expression profile.In one embodiment, with the form of compound decision-making scoring, summarized gene expression profile from one group of transcript of patient tissue sample, and with the contrast threshold value relatively, described contrast threshold value for example learns from the set of patient data the threshold value that derives.Described threshold value is separated patient's group based on different characteristic, and described feature is such as but not limited to good/poor prognosis, to the response for the treatment of/non-response, cancer detection/diagnosis and cancer classification.Patient's set certificate is preferably derived from colon's sample, and described colon sample responds to characterize by prognosis, recurrence possibility or long-term surviving, diagnosis, cancer classification, personalized genome spectrum, clinical effectiveness, treatment.The feature of the clinical samples in the training group of the decision-making value homonymy that can derive with mathematics from the express spectra of clinical samples and corresponding decision-making scoring is associated.In this embodiment, optimize the threshold value of linear classification compound decision-making scoring so that the sensitivity according to cross validation of application and the maximization of specific summation in the training data group.These methods also are used for the prognosis of determining colon cancer and suffering from the patient of II phase colon cancer a specific embodiments.In some instances, the clinical effectiveness of disclosed method predicted difference, it can for example be measured in the following areas: for example after the exenterate cancer, and perhaps after the exenterate cancer with the NACT combination, the cancer return risk of the survival of shortening or increase.

The method of diagnosis available from colon cancer in experimenter's sample is provided.This method comprises the expression that detects available from the listed at least two kinds of colon cancer associated nucleic acid molecules of table 6 in experimenter's the sample that comprises nucleic acid, and the expression of more described at least two kinds of colon cancer associated nucleic acid molecules or from the contrast threshold value of the decision-making of its derivation scoring with the indication diagnosis of colon cancer, the wherein expression of threshold value homonymy or from the diagnosis of the decision-making scoring indication colon cancer of its derivation.In some instances, the contrast threshold value is the threshold value that always the corresponding transcript of the listed colon cancer associated nucleic acid of table 6 molecule is derived in known colon cancer sample (or a plurality of sample).

The method that is used for classification colon cancer sample is provided.This method comprises the expression that detects available from the listed at least two kinds of colon cancer associated nucleic acid molecules of table 6 in experimenter's the sample that comprises nucleic acid, and the expression of more described at least two kinds of colon cancer associated nucleic acid molecules or from the contrast threshold value of the decision-making of its derivation scoring with the known classification of indication, wherein at the expression of described threshold value homonymy or allow the classification of described colon cancer sample from the decision-making scoring of its derivation.In some instances, the contrast threshold value is the threshold value that always the corresponding transcript of the listed colon cancer associated nucleic acid of table 6 molecule is derived in the colon cancer sample (or a plurality of sample) of known classification.In some instances, the colon cancer sample is classified as I phase, II phase, III phase and IV phase.In some instances, described method also comprises that selection will be for the effective treatment plan of the colon cancer of classifying, for example excision, chemotherapy, radiation or its combination in any.

Provide and be used for prediction to colon cancer, for example suffered from the experimenter's of II phase colon cancer the method for response for the treatment of.This method comprises the expression that detects available from least two kinds of listed colon cancer associated nucleic acid molecules of table 6 in experimenter's the sample that comprises nucleic acid, and the expression of more described at least two kinds of colon cancer associated nucleic acid molecules or from the contrast threshold value of the known response to treatment of the decision-making of its derivation scoring and indication, wherein at the expression of described threshold value homonymy or from the similar response to treatment of decision-making scoring indication of its derivation, thereby prediction is to the response for the treatment of.In some instances, the contrast threshold value is the threshold value of the corresponding transcript derivation of the listed colon cancer associated nucleic acid of table 6 molecule from the colon cancer sample (or a plurality of sample) from having the known response to treatment.In some embodiments, described method is the method for the response of prediction excision, chemotherapy, radiation or its combination in any.

Provide to be used for prediction and to suffer from the experimenter of colon cancer, for example be diagnosed with the experimenter's of II phase colon cancer the method for long-term surviving.These methods comprise the expression that detects available from least two kinds of listed colon cancer associated nucleic acid molecules of table 6 in experimenter's the sample that comprises nucleic acid, and the expression of more described at least two kinds of colon cancer associated nucleic acid molecules or have the contrast threshold value of long-term surviving history from the decision-making of its derivation scoring and indication, wherein at the expression of described threshold value homonymy or from the described experimenter's of decision-making scoring indication of its derivation long-term surviving, thus prediction experimenter's long-term surviving.In some instances, the contrast threshold value is the threshold value of the corresponding transcript derivation of the listed colon cancer associated nucleic acid of table 6 molecule from the experimenter available from having long-term surviving history (or a plurality of experimenter's) colon cancer sample (or a plurality of sample).

Also provide and be used for the prediction experimenter, for example be diagnosed with the method for colon cancer recurrence in the experimenter of II phase colon cancer.These methods comprise the expression that detects available from least two kinds of listed colon cancer associated nucleic acid molecules of table 6 in experimenter's the sample that comprises nucleic acid, and the expression of more described at least two kinds of colon cancer associated nucleic acid molecules or from the historical contrast threshold value of the decision-making of its derivation scoring and indication recurrence, wherein in expression or the recurrence from the described experimenter of decision-making scoring indication of its derivation of described threshold value homonymy.In some instances, the contrast threshold value is the threshold value that derives from the corresponding transcript from having the listed colon cancer associated nucleic acid of table 6 molecule the historical colon cancer sample (or a plurality of sample) of recurrence.

Method for the preparation of the personalized colon cancer genome spectrum of experimenter is provided.These methods comprise the expression that detects available from least two kinds of listed colon cancer associated nucleic acid molecules of table 6 in experimenter's the sample that comprises nucleic acid, and produce the report of summarizing the data that obtain by gene expression analysis.

in the particular of method disclosed herein, measure table 6 transcription thing or its expression product at least 2, for example at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 11, at least 12, at least 13, at least 14, at least 15, at least 20, at least 25, at least 30, at least 35, at least 40, at least 45, at least 50, at least 55, at least 60, at least 65, at least 70, at least 75, at least 80, at least 85, at least 90, at least 95, at least 100, at least 125, at least 150, at least 175, at least 200, at least 225, at least 250, at least 275, at least 300, at least 325, at least 350, at least 375, at least 400, at least 425, at least 450, at least 474, at least 500, at least 525, at least 550, at least 575, at least 600, at least 634 or the expression of even whole 636 and with the contrast threshold value relatively.In other embodiments of these methods, measure the expression of MUM1 and SIGMAR1 or its expression product and with the contrast threshold value, compare.In another embodiment, measure MUM1, SIGMAR1, ARSD, SULT1C2 and PPFIBP1 or its expression product expression and with the contrast threshold value relatively.In other embodiments, measure the antisense sequences of ARSD, CXCL9, PCLO, SLC2A3, FCGBP, SLC2A14, SLC2A3, BCL9L and MUC3A, OLFM4 and RNF39 or its expression product expression and with the contrast threshold value relatively.In other embodiments, in Simultaneous Determination table 1, one of 2 and/or 6 listed basically all the expression of transcripts and with the contrast threshold value relatively.

In some embodiments of disclosed method, according to the variability of the quality of the RNA of the difference of the amount of the RNA that measures and use, proofread and correct (standardization) rna level.Control transcripts can be used as positive or negative contrast and is included in mensuration, and standardization reading and guarantee reliable measurement data, but preferably is omitted when carrying out actual prognosis.The former actual identity is usually inessential, and can imagine various transcripts for all purposes disclosed herein., for standardization contrast, it is contemplated that various transcripts, but they must meet between various experimenters or interested target tissue illness, the basic demand of constant and stable expression between the prognosis group in particularly considering.Similarly, RNA degraded contrast must show the intensity behavior of the RNA that is fit to indication (extremely) degraded.This can comprise or can not comprise RNA contrast, and it shows stable intensity, no matter and as the general RNA degraded of the sample of positive control.About these contrasts, the intensity mode of other RNA contrasts that will analyze be fit to, observe intensity dependence to the RNA degradative phase to it.This can comprise or can not comprise specific analysis, depends on the diverse location of probe sequence with respect to transcript 3 ' end.

Microarray is used for some embodiments of the disclosure method of quantitate gene expression therein, can use one or more of following contrast:

(a) comparison contrast, it is the specific transcript that adds with mark pattern, suitable Grid Align during its image that is bonded to the ad-hoc location on array and guarantees scanning array is processed.

(b) amplification contrast, it is the specific unlabelled transcript that added before carrying out any amplification, for example the polyadenylic acid control transcripts, therefore experienced the processing identical with sample mRNA, to guarantee synthetic and suitably the carrying out of amplified reaction subsequently of cDNA.

(c) mark and hybridization contrast, it is at mark and hybridizes to the specific contrast of adding before chip, amplified reaction before being used for being independent of and control the efficiency of this two step.

(d) background contrast, it is the probe sequence that does not have in sample on microarray that corresponding target sequence should obtain.Therefore, there is no in principle specific target in conjunction with occurring.These contrasts are used for setting up background or crisscrossing intensity.They may characterize by GC content different on whole microarray and suitable space distribution.

(e) standardization contrast, it is the probe sequence that detects from the special target sequence of selecting of sample, is used for proofreading and correct different mRNA input quantities, different amplified reaction productive rate and the different overall sensitivity of measurement mechanism.They are used for the intensity level of correcting measuring, and therefore guarantee the analysis precision of the increase of whole measurement mechanism, comprise the preparing experiment step.

(f) RNA quality and degraded contrast, it is its probe sequence of diverse location of 3 ' position of gene separately that comes from respect to being designed to indicate the RNA quality and detecting the RNA degraded.May represent different RNA degradation behavior from the different RNA kind from the correspondent probe of a plurality of genes or probe groups.

And a) – d of contrast) can consider and derive based on sequence purely, and should not naturally be present in interested tissue and illness, contrast e) and f) can select by the suitable analysis of patient data before.This can be or can not be the identical training data of prognosis gene signature derivative according to it.

Should be appreciated that, above-mentioned contrast only provides as an example, and it is contemplated that other embodiments of the present disclosure (for example qPCR), wherein use is had the difference contrast of identity function.

B. probe, primer and array

Probe and the primer special to disclosed colon cancer gene signature are disclosed.Also disclose array, comprised the probe of disclosed colon cancer signature.In some embodiments, the special probe of disclosed colon cancer gene signature is comprised nucleotide sequence with one of SEQ ID NO:1-636 or its complement specific hybrid.in some embodiments, the probe groups of disclosed colon cancer signature comprises at least 2 with table 6 transcription thing, for example at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 11, at least 12, at least 13, at least 14, at least 15, at least 20, at least 25, at least 30, at least 35, at least 40, at least 45, at least 50, at least 55, at least 60, at least 65, at least 70, at least 75, at least 80, at least 85, at least 90, at least 95, at least 100, at least 125, at least 150, at least 175, at least 200, at least 225, at least 250, at least 275, at least 300, at least 325, at least 350, at least 375, at least 400, at least 425, at least 450, at least 474, at least 500, at least 525, at least 550, at least 575, at least 600, at least 634 or the probe of even whole 636 specific hybrids, described transcript has the weight limit that is defined as the grade of average weight in the compound decision-making score function of measuring according to cross validation, and still has prognosis values, for example with any one of SEQ ID NO:1-636 or the probe of its complement specific hybrid.In some embodiments, the probe groups of disclosed colon cancer signature comprises front 10 the weighting transcripts listed with table 6, the probe of second front 10 weighting transcripts, the 3rd front 10 weighting transcripts, the 4th front 10 weighting transcripts, the 5th front 10 weighting transcripts, the 6th front 10 weighting transcripts, the 7th front 10 weighting transcripts, the 8th front 10 weighting transcripts, the 9th front 10 weighting transcripts or the tenth front 10 weighting transcript specific hybrids.and in other embodiments, the probe groups of disclosed colon cancer signature comprises with table 6 is listed having 636 of a weight limit, 634, 620, 610, 600, 590, 580, 570, 560, 550, 540, 530, 520, 510, 500, 490, 480, 470, 460, 450, 440, 430, 420, 410, 400, 390, 380, 370, 360, 350, 340, 330, 320, 310, 300, 290, 280, 270, 260, 250, 240, 230, 220, 210, 200, 190, 180, 170, 160, 150, 140, 130, 120, 110, 100, 90, 80, 70, 60, 50, 40, 30, the probe of 20 or 10 transcripts or its complement specific hybrid.In some embodiments, the probe groups of disclosed colon cancer signature comprises approximately 200 to about 1000 probes, for example approximately 400 to about 800 probes, for example approximately 500 to about 700 probes, for example approximately 550 to approximately 650 probes, the wherein transcript of probe in detecting table 6.Extra probe can randomly be selected from those that detect the transcript of expressing in colon cancer, perhaps as signal contrast or expression contrast and play a role.The examples of such optional probe can be selected from Colorectal Cancer DSA ^TMThose that comprise on instrument.

In some embodiments, the probe groups of disclosed colon cancer signature comprises the probe with the transcript specific hybrid of MUM1 and SIGMAR1.In other embodiments, the probe groups of disclosed colon cancer signature comprises the probe with the transcript specific hybrid of MUM1, SIGMAR1, ARSD, SULT1C2 and PPFIBP1.And in other embodiments, the probe groups of disclosed colon cancer signature comprises the probe with the transcript specific hybrid of the antisense sequences of the transcript of ARSD, CXCL9, PCLO, SLC2A3, FCGBP, SLC2A14, SLC2A3, BCL9L and MUC3A, OLFM4 and RNF39.Can prepare the one group of probe or the primer that basically represent gene expression signature." basically represent gene expression signature " and refer to gene expression signature in coding or non-coding transcript at least 50%, 60%, 70%, 80%, 90%, 95%, 99% or 100%, the probe groups of at least 50%, 60%, 70%, 80%, 90%, 95%, 99% or 100% specific hybrid of coding or non-coding transcript or its complement in the gene expression signature shown in table 1,2 or 6 for example.

Advantageously using the probe with 3 ' regional combination of gene expression signature transcription thing, is particularly from the situation of the RNA of paraffin-embedded tissue extraction at the patient tissue of gene expression to be analyzed.Usually, each probe can with transcript separately in complementary sequence hybridization, it appears in the 1kb or 500bp or 300bp or 200bp or 100bp of transcript 3 ' end.In the mRNA situation, " 3 ' end of transcript " is defined as the polyadenylation site at this paper, do not comprise poly-A tail.

In one embodiment, use to form 30% probe set (pool) of total absolute weight of signature.In optional embodiment, 40%, 60%, 70%, 80%, 90%, 95% or 100% of total absolute weight of use formation signature probe set in method described herein.The basis of including mark in and the clinical importance that changes with respect to the mRNA level of reference group have below been indicated.In some embodiments, disclosed probe is the part of array, and for example, probe is bonded to solid substrate.The method of exemplary nucleic acid array and this array of preparation has been discussed in following chapters and sections D.

In some embodiments, be nucleic acid array to the special probe of disclosed colon cancer gene signature, the part of microarray for example.In some instances, this array comprises the nucleotide sequence with one of SEQ IDNO:1-636 or its complement specific hybrid.in some embodiments, nucleic acid array, for example microarray comprises with table 6 transcription thing at least 2, for example at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 11, at least 12, at least 13, at least 14, at least 15, at least 20, at least 25, at least 30, at least 35, at least 40, at least 45, at least 50, at least 55, at least 60, at least 65, at least 70, at least 75, at least 80, at least 85, at least 90, at least 95, at least 100, at least 125, at least 150, at least 175, at least 200, at least 225, at least 250, at least 275, at least 300, at least 325, at least 350, at least 375, at least 400, at least 425, at least 450, at least 474, at least 500, at least 525, at least 550, at least 575, at least 600, at least 634 or the probe of even whole 636 specific hybrids.In some embodiments, the nucleic acid array of disclosed colon cancer signature comprises front 10 the weighting transcripts listed with table 6, the probe of second front 10 weighting transcripts, the 3rd front 10 weighting transcripts, the 4th front 10 weighting transcripts, the 5th front 10 weighting transcripts, the 6th front 10 weighting transcripts, the 7th front 10 weighting transcripts, the 8th front 10 weighting transcripts, the 9th front 10 weighting transcripts or the tenth front 10 weighting transcript specific hybrids.and in other embodiments, comprise with table 6 is listed having 636 of a weight limit for the nucleic acid array of disclosed colon cancer signature, 634, 620, 610, 600, 590, 580, 570, 560, 550, 540, 530, 520, 510, 500, 490, 480, 470, 460, 450, 440, 430, 420, 410, 400, 390, 380, 370, 360, 350, 340, 330, 320, 310, 300, 290, 280, 270, 260, 250, 240, 230, 220, 210, 200, 190, 180, 170, 160, 150, 140, 130, 120, 110, 100, 90, 80, 70, 60, 50, 40, 30, the probe of 20 or 10 transcripts or its complement specific hybrid.In some embodiments, the nucleic acid array of disclosed colon cancer signature comprises approximately 200 to about 1000 probes, for example approximately 400 to about 800 probes, for example approximately 500 to about 700 probes, for example approximately 550 to approximately 650 probes, the wherein transcript of probe in detecting table 6.Extra probe can randomly be selected from those that detect the transcript of expressing in colon cancer, perhaps its as signal contrast or expression contrast and play a role.The probe of examples of such optional can be selected from Colorectal Cancer DSA ^TMThose that comprise on instrument.In some embodiments, the nucleic acid array of disclosed colon cancer signature comprises and surpasses approximately 1000 probes.

The primer pair of the gene expression signature that is used for amplification colon cancer nucleic acid is also disclosed.In some instances, primer pair comprise length be 15 to 40 nucleotide, comprise the forward primer with the nucleotide sequence of any one or its complement specific hybrid of nucleotide sequence shown in SEQ ID NO:1-636, with length be 15 to 40 nucleotide, comprise and the reverse primer of the nucleotide sequence of any one or its complement specific hybrid of nucleotide sequence shown in SEQ ID NO:1-636, wherein primer sets can instruct the amplification of nucleic acid.

The primer pair group of the gene expression signature that is used for amplification colon cancer nucleic acid is also disclosed.in some embodiments, the primer sets of disclosed colon cancer signature comprises at least 2 with the listed transcript of table 6, for example at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 11, at least 12, at least 13, at least 14, at least 15, at least 20, at least 25, at least 30, at least 35, at least 40, at least 45, at least 50, at least 55, at least 60, at least 65, at least 70, at least 75, at least 80, at least 85, at least 90, at least 95, at least 100, at least 125, at least 150, at least 175, at least 200, at least 225, at least 250, at least 275, at least 300, at least 325, at least 350, at least 375, at least 400, at least 425, at least 450, at least 474, at least 500, at least 525, at least 550, at least 575, at least 600, at least 634 or even whole 636 specific hybrids and its primer of can increasing, described transcript has the weight limit that is defined as the grade of average weight in the compound decision-making score function of measuring according to cross validation, and still has prognosis values, for example with any one or its complement specific hybrid of SEQ ID NO:1-636 and its primer of can increasing.In some embodiments, the primer sets of disclosed colon cancer signature comprises front 10 the weighting transcripts listed with table 6, second front 10 weighting transcripts, the 3rd front 10 weighting transcripts, the 4th front 10 weighting transcripts, the 5th front 10 weighting transcripts, the 6th front 10 weighting transcripts, the 7th front 10 weighting transcripts, the 8th front 10 weighting transcripts, the 9th front 10 weighting transcripts or the tenth front 10 weighting transcript specific hybrids and its primer of can increasing.and in other embodiments, the primer sets of disclosed colon cancer signature comprises with table 6 is listed having 636 of a weight limit, 634, 620, 610, 600, 590, 580, 570, 560, 550, 540, 530, 520, 510, 500, 490, 480, 470, 460, 450, 440, 430, 420, 410, 400, 390, 380, 370, 360, 350, 340, 330, 320, 310, 300, 290, 280, 270, 260, 250, 240, 230, 220, 210, 200, 190, 180, 170, 160, 150, 140, 130, 120, 110, 100, 90, 80, 70, 60, 50, 40, 30, 20 or 10 transcripts or its complement specific hybrid and its primer of can increasing.

In some embodiments, the primer sets of disclosed colon cancer signature comprises and the transcript specific hybrid of MUM1 and SIGMAR1 and its primer of can increasing.In another embodiment, the primer sets of disclosed colon cancer signature comprises and the transcript specific hybrid of MUM1, SIGMAR1, ARSD, SULT1C2 and PPFIBP1 and its primer of can increasing.And in another embodiment, the probe groups of disclosed colon cancer signature comprises the probe with the transcript specific hybrid of the antisense sequences of the transcript of ARSD, CXCL9, PCLO, SLC2A3, FCGBP, SLC2A14, SLC2A3, BCL9L and MUC3A, OLFM4 and RNF39.Can prepare the one group of probe or the primer that basically represent gene expression signature." basically represent gene expression signature " and refer to gene expression signature in coding or non-coding transcript at least 50%, 60%, 70%, 80%, 90%, 95%, 99% or 100%, the probe groups of at least 50%, 60%, 70%, 80%, 90%, 95%, 99% or 100% specific hybrid of coding or non-coding transcript or its complement in the gene expression signature shown in table 1,2 or 6 for example.

C. the statistical survey of colon cancer label

Can be by the disclosed colon cancer signature of evaluation of statistical methods.In some embodiments, the gene expression profile by linear classifier evaluate patient tissue samples.As used herein, linear classifier refers to that the genes of individuals intensity weighted sues for peace into compound decision-making scoring (" decision function ").Then the decision-making scoring is held back threshold value relatively with predetermined, be scheduled to hold back threshold value corresponding to sensitivity and specific certain set point, its indication sample is higher than threshold value (decision function is positive) or lower than threshold value (decision function is negative).

Effectively, this means data space, that is, the gene expression value likely the group of combination be divided into two half that mutually repel, corresponding to different clinical classifications or prediction, for example, one corresponding to good prognosis and another is corresponding to poor prognosis.In the context of whole signature, relative cross to express of some gene can increase decision-making scoring (positive weight) or reduce decision-making scoring (negative weight), and therefore promotes for example aggregate decision of poor or good prognosis.

The explanation of this amount, that is, and good holding back the period of expansion (" training ") that threshold value comes from one group of patient with known results poor prognosis.The respective weights of decision-making scoring is become reconciled/differ from, and to hold back threshold value definite from training data priori by method known to persons of ordinary skill in the art in prognosis.In a preferred embodiment of this method, use partial least squares discriminant analysis (PLS-DA) determine weight (

J.Chemom.1185-196,1987; Nguyen and Rocke, Bioinformatics18 39-50,2002).When being used for the transcript of colon cancer signature, additive method of classifying well known by persons skilled in the art also can be together with method described herein.

Can use diverse ways will change into for the quantitative data of these genes or the measurement of its product prognosis or other prediction purposes.These methods include but not limited to pattern-recognition (the people Pattern Classification such as Duda, the 2nd edition, John Wiley, New York2001), machine learning (

Deng people Learning with Kernels, MIT Press, Cambridge2002, Bishop, Neural Networks for Pattern Recognition, Clarendon Press, Oxford1995), statistics (the people The Elements of Statistical Learning such as Hastie, Springer, New York2001), the bioinformatics (people such as Dudoit, J.Am.Statist.Assoc.97:77-87,2002; The people such as Tibshirani, Proc.Natl.Acad.Sci.USA99:6567-6572,2002) or chemometrics (Vandeginste, Deng the people, Handbook of Chemometrics and Qualimetrics, part B, Elsevier, Amsterdam1998).

In some embodiments, in training step, measure one group of clinical samples of good and poor prognosis situation, and use intrinsic information Optimization Prediction method from this training data with optimum prediction training group or following sample group.In this training step, the method that training or parametrization are used is with the certain strength model prediction from for specific prognosis conversation (prognostic call).Before experience method of prognosis or algorithm, conversion or the pre-treatment step that can be fit to the data of measurement.

In some embodiments, form each transcript the pre-service intensity level the weighting sum and with threshold value for the training group relatively (the people Pattern Classification such as Duda, the 2nd edition, John Wiley, New York2001).Can derive weights by a large amount of linear classification methods, these methods include but not limited to partial least square method (PLS, people such as (, 2002, Bioinformatics18 (2002) 39-50) Nguyen) or support vector machine (SVM, (

Deng people Learning with Kernels, MIT Press, Cambridge2002)).

In some embodiments, before application weighting sum, data are non-linearly transformed, for example described above.This non-linear conversion can comprise the dimension that increases data.Non-linear conversion and weighted sum can also for example be carried out by using kernel function implicitly.(

Deng people Learning with Kernels, MIT Press, Cambridge2002).

In some embodiments, new data sample and two or more kind prototype relatively, are training sample or the artificial prototypes that produces that reality is measured.This relatively uses suitable similarity measurement to carry out, such as but not limited to Euclidean distance (the people Pattern Classification such as Duda, the 2nd edition, John Wiley, New York2001), related coefficient (van ' t Veer, Deng the people, Nature415:530,2002) etc.Then, new samples is appointed as is had Nearest prototype or the group of the prognosis near prototype of high number.

In some embodiments, use decision tree (the people The Elements of Statistical Learning such as Hastie, Springer, New York2001) or random forest (Breiman, 2001Random Forests, Machine Learning45:5) make the prognosis conversation from the intensity data of the measurement of transcribing group or its product.

In some embodiments, use neural network (Bishop, Neural Networks for Pattern Recognition, Clarendon Press, Oxford1995) to make the prognosis conversation from the intensity data of the measurement of transcribing group or its product.

In some embodiments, use discriminatory analysis (the people Pattern Classification such as Duda, the 2nd edition, John Wiley, New York2001) make the prognosis conversation from the intensity data of transcribing group or the measurement of its product, described discriminatory analysis includes but not limited to linearity, Diagonal Linear, quadratic term and logic discrimination analysis.

In some embodiments, the forecast analysis (PAM, people such as (, Proc.Natl.Acad.Sci.USA99:6567-6572,2002) Tibshirani) of using microarray is made the prognosis conversation from the intensity data of the measurement of transcribing group or its product.

In some embodiments, the soft Independent modeling (SIMCA, (Wold, 1976, Pattern Recogn.8:127-139)) of use classes simulation is made the prognosis conversation from the intensity data of the measurement of transcribing group or its product.

The detection method of D.mRNA

Can estimate gene expression by the encode mRNA of gene of interest of detection.Therefore, disclosed method can comprise evaluation mRNA.Can use method known to persons of ordinary skill in the art, comprise the obtainable kit of commercial sources from the tumor sample from the experimenter, from experimenter's adjacent nonneoplastic tissue sample, from normal (health) experimenter without tumor tissues sample or its combination isolation of RNA.

The conventional method that is used for the mRNA extraction is well known in the art, and is disclosed in molecular biological national textbook, comprises the people such as Ausubel, Current Protocols of Molecular Biology, John Wiley and Sons (1997).The method that is used for extracting from paraffin-embedded tissue RNA is disclosed in for example Rupp and Locker, Biotechniques6:56-60, and 1988, and the people such as De Andres, Biotechniques18:42-44,1995.In an example, can use purification kit, buffering agent group and proteinase from the commodity production business, carry out RNA according to manufacturer's instructions and separate, described manufacturer for example

(Valencia, CA).For example, can use

Mini post separates total RNA of cell in culture (for example available from experimenter those).The obtainable RNA separating kit of other commercial sources comprises

Global DNA and RNA purification kit ( Madison, Wis.), and paraffin mass RNA separating kit (Ambion, Inc.).Can use the total RNA of RNA Stat-60 (Tel-Test) separation from tissue samples.Can for example by cesium chloride density gradient centrifugation, separate from the RNA of tumour or the preparation of other biological sample.

The use of the paraffin embedding biopsy material of the file of the mensuration of all marks in signature described herein and method adaptation group, and therefore compatible with the biopsy material of the most general available types.Can use available from formalin fixing, paraffin-embedded tissue samples, fresh food frozen tissue or for example be stored in solution In the RNA of flesh tissue measure the expression of colon's sample transcription thing.For example, can after any one of said procedure or during whole application, perhaps by any other method known in the art, carry out the separation of RNA., although any technology of gene expression profile and protein technique are suitable for carrying out method described herein, often by the DNA microarray technology, measure gene expression dose.

If tissue-derived is that formalin is fixing, paraffin-embedded tissue samples, RNA may, by fragmentation, cause information dropout.Signature provided herein is derived from from the transcript set of its 3 ' end order-checking, thereby the accurate displaying of transcribing group of tissue is provided.Therefore, signature provided herein is used for fresh food frozen and fixing paraffin-embedded tissue.

In some embodiments, the RNA sample that is used for method described herein can be from fixing, paraffin-embedded colon sample preparation, and by using the one or more of following steps, for example following steps is whole:

(a) use conventional method and in organic solvent with a plurality of washing step deparaffnizes;

(b) air dry and with Protease Treatment with in broken cells and the iuntercellular key, cause RNA to discharge from tissue;

(c) remove the genomic DNA that pollutes;

(d) wash in organic solvent; And be fit to without RNA enzyme elution buffer in wash-out.

The RNA extracting method can also be included in height sex change lysis buffer and hatch tissue, described height sex change lysis buffer has to reverse preserves the crosslinked additional functionality of formalin that occurs in tissue, in this mode as measured downstream performance improvement RNA productive rate and quality.

After RNA reclaims, can randomly be further purified RNA, obtain being substantially free of the DNA of pollution or the RNA of protein.Any one of the aforementioned techniques that can reclaim by RNA or for example use the obtainable RNA scavenger reagent of commercial sources box

The scavenger reagent box (

) complete further RNA purifying.For example, tissue specimen can be available from tumour, and RNA can be available from the microdissection part of the tissue specimen of tumour cell enrichment.

The method of gene expression profile comprises the method for analyzing based on multi-nucleotide hybrid and the method that checks order based on polynucleotide.In some instances, make with the following method mrna expression in quantitative sample: Northern trace or in situ hybridization (Parker﹠amp; Barnes, Methods in Molecular Biology106:247-283,1999); The RNA enzyme protection is measured (Hod, Biotechniques13:852-4,1992); With the method for PCR-based, such as reverse transcriptase polymerase chain reaction (RT-PCR) (people such as Weis, Trends in Genetics8:263-4,1992).Alternatively, can adopt the antibody that can identify specific duplex, described duplex comprises DNA duplex, RNA duplex and DNA-RNA heteroduplex body or DNA-protein duplex.Comprise the continuous analysis (SAGE) of gene expression and by the sign gene expression analysis of order-checking (MPSS) of large-scale parallel based on the method that represents of gene expression analysis of order-checking.In an example, can use the mRNA level in the more different samples of RT-PCR,, to characterize gene expression pattern, distinguish closely related mRNA and analyze the RNA structure.In specific sample, by nucleic acid microarray technology, round pcr or its, make up to analyze disclosed colon cancer signature.

1. use the gene expression profile of microarray method

In some embodiments, colon cancer related gene and/or transcript, the express spectra of those that example is as shown in table 6 can use microarray technology to measure in fresh or paraffin-embedded tumor tissues.In the method, interested polynucleotide sequence, for example with the polynucleotide sequence of nucleotide sequence shown in table 6 or its complement specific hybrid by bed board or be arranged in the microchip substrate.Then, the sequence of arrangement with from the nucleic acid hybridization of interested cell or tissue.

In RT-PCR method (vide infra), total RNA that normally separate with corresponding normal structure or clone from people's tumour or tumor cell line in the source of mRNA.Therefore, RNA can separate from multiple primary tumor or tumor cell line.If the source of mRNA is primary tumor, for example, mRNA can extract from the paraffin embedding of freezing or file and/or fixing (for example formalin is fixed) tissue samples, its preparation and preservation in the clinical practice of every day routinely.

In the particular of microarray technology, the embolus of the pcr amplification of cDNA clone or oligonucleotides is applied to substrate in closely spaced array.Can also use for example photoetching process and being combined in substrate of solid state chemistry synthetic technology of based semiconductor directly to synthesize short oligonucleotides.(Affymetrix，Inc.，Santa?Clara，CA)。In one embodiment, at least 10,000 nucleotide sequence is present in substrate.Microarray transcript fixing in substrate is fit to hybridize under stringent condition.Fluorescently-labeled nucleotide probe can be incorporated fluorescent nucleotide into by the reverse transcription that extracts the RNA that knits from interest groups and produce.Be applied to the label probe of array and each the nucleotide specific hybrid on array.After the probe of non-specific binding is removed in washing, by the confocal laser microarray or by another detection method for example the CCD camera carry out scanning array.The corresponding transcript abundance of quantitative permission assessment of the hybridization of the element of each arrangement.

Use dual base color fluorescence, the nucleotide probe of the separate marking that produces from two sources can be in couples and hybridization array.The hybridization scale of miniaturization provides the convenience of lots of genes expression pattern and has estimated fast.Shown this method have detect the required sensitivity of the rare transcript of with each cell minority copy, expressing and duplicate detection expression at least about the required sensitivity of twice difference (people such as Schena, Proc.Natl.Acad.Sci.USA93 (2): 106149 (1996)).Can also carry out microarray analysis by the obtainable equipment of commercial sources according to manufacturer's scheme, for example by using Affymetrix

Technology (Affymetrix, Inc., Santa Clara, CA) or Agilent microarray technology (Agilent Technologies, Inc., Santa Clara, CA).

The development of the microarray method of large scale analysis gene expression make the molecular marker of system retrieval cancer classification and kinds of tumors type for example the prediction of result in the colon cancer tumour become possibility.

In particular provided herein, can use array to estimate the colon cancer gene expression profile, for example with prognosis or diagnosis, suffer from the patient of colon cancer.When describing mainly the array that is formed by the special probe of his-and-hers watches 1, the listed gene of table 2 and/or the listed transcript of table 6 or primer, this array comprises probe or the primer special to these colon cancer related genes, and may further include contrast probe (for example, to confirm that incubation conditions is enough).Exemplary contrast probe comprises GAPDH, beta-actin and 18S RNA.

I. array base palte

The solid support of array can be formed by inorganic material (for example glass) or organic polymer.The material that is fit to solid support includes but not limited to: the potpourri of the biaxially oriented polypropylene of polypropylene, tygon, polybutylene, polyisobutylene, polybutadiene, polyisoprene, polyvinylpyrrolidine, teflon, polyvinylidene fluoride, polyfluoroethylene-propylene, poly ethylene vinyl alcohol, polymethylpentene, polychlorotrifluoroethylene, polysulfones, hydroxylated biaxially oriented polypropylene, ammonification, biaxially oriented polypropylene, ethylene acrylic, ethylene methacrylic acid and the multipolymer thereof of mercaptan is (referring to U.S. Patent number 5,985,567).

Generally speaking, the suitable feature that can be used to form the material of solid support surface comprises:, in compliance with surface active, make after activation, the surface of holder can covalently bound biomolecule, and for example oligonucleotides is on it; The biddability of " original position " synthetic biological molecule; Chemically inert, make the zone that by oligonucleotides, is not occupied on holder not be obedient to non-specific binding, perhaps when non-specific binding occurred, this material can easily not removed oligonucleotides from surface removal.

In another example, the organic polymer of surface active is as solid support surface.An example of the organic polymer of surface active is the polypropylene material by the RF Plasma Discharge ammonification.Can also use other reactive groups, for example carboxylation, hydroxylation, mercaptan or active ester group.

Ii. array format

Can adopt various array formats according to the disclosure.An example comprises the linear array of oligonucleotides key, generally in this area, is called dipstick.Another form that is fit to comprises the two-dimensional model (for example, 4096 squares in 64 * 64 arrays) of separate unit.As understood by a person skilled in the art, other array formats are also applicable on an equal basis, include but not limited to crack (rectangle) and annular array (referring to U.S. Patent number 5,981,185).In some instances, array is porous plate.In an example, array forms on polymeric media, and it is helical, film or film.An example of organic polymer medium is the extremely approximately polypropylene of 20 mils of about 1 mil of thickness (0.001 inch), but film thickness is not crucial, and can in very large range change.Array can comprise biaxially oriented polypropylene (BOPP) film, and except its persistence, it also shows low background fluorescence.

Array format of the present disclosure can be included in the form of number of different types." form " comprises that solid support can be attached to its any form, for example microtiter plate (for example, porous plate), test tube, inorganic sheet, dipstick etc.For example, when solid support was the polypropylene helical, one or more polypropylene helicals can be attached to plastics dipstick type device; Polypropylene screen can be attached to microslide.Concrete form itself is unessential.Be necessary that solid support can add on it and do not affect solid support or the behaviour of any XC polymer of absorbing on it, and any material (for example, clinical sample and hybridization solution) that form (for example dipstick or microslide) is introduced for device is stable.

Array of the present disclosure can prepare by several different methods.In an example, oligonucleotides or protein sequence are synthesized separately, then with solid support, are connected (referring to U.S. Patent number 6,013,789).In another example, sequence directly is blended on support so that required array (referring to U.S. Patent number 5,554,501) to be provided.Be fit to oligonucleotides and protein is covalently coupled to solid support and directly synthetic oligonucleotide or method of protein are well known by persons skilled in the art on holder; The general introduction of proper method can be referring to people such as Matson, Anal.Biochem.217:306-10,1994.In an example, use the conventional chemical technology for preparing oligonucleotides on solid support that oligonucleotides is blended into holder (for example PCT applies for WO85/01051 and WO89/10977, or U.S. Patent number 5,554,501).

Can use by the automation equipment generation suitable array of the precursor with four bases of preassigned pattern layout with synthetic oligonucleotide in array element.In brief, adopt the chemical delivery system of hyperchannel robotization to produce oligonucleotide probe colony (number of active lanes in quantitatively corresponding to delivery system) with parallel columns in whole substrate.First direction complete oligonucleotides synthetic after, can the substrate half-twist is synthetic to proceed at present vertical with first group second group is listed as to allow.This process produces multichannel array, and its intersection produces a plurality of separate units.

Oligonucleotides can or be held and be bonded to the polypropylene support by 5 ' of oligonucleotides by 3 ' end of oligonucleotides.In an example, oligonucleotides is bonded to solid support by 3 ' end.Yet those skilled in the art can determine whether 3 ' end of oligonucleotides or the use of 5 ' end are fit to be bonded to solid support.Generally speaking, the complementary decision in inside of oligonucleotide probe and the combination of holder in the zone of 3 ' end and 5 ' end.

In particular instance, the oligonucleotide probe on array comprises that permission detects oligonucleotide probe: one or more marks of target sequence hybridization complex.

2. use the gene expression profile of microarray method

The sensitiveest and the most flexibly one of quantivative approach be RT-PCR, it can be used for relatively being with or without the mRNA level in different sample colony in the normal and tumor tissues of drug therapy, to characterize gene expression pattern, to distinguish closely-related mRNA and to analyze the RNA structure.

The first step is from target sample for example people's tumour or tumor cell line and corresponding normal structure or clone isolation of RNA separately., if the source of RNA is primary tumor, can for example from the paraffin embedding of freezing or file and/or fixing (for example formalin is fixed) tissue samples, extract RNA.

The variation of RT-PCR is real-time quantitative RT-PCR, and its fluorescence probe by double labeling (for example,

Probe) measure the accumulation of PCR product.The internal competition thing of PCR in real time and each target sequence wherein be used for standardized quantitative competitive PCR and use standardization gene that sample contains or the quantitative comparison PCR of the house-keeping gene of RT-PCR compatible (referring to people such as Heid, Genome Research6:986-994,1996).Quantitative PCR also is described in U.S. Patent number 5,538,848.Relevant probe and quantitative amplification program description be in U.S. Patent number 5,716, and 784 and U.S. Patent number 5,723,591.Can be available from PE Applied Biosystems (Foster City, CA) for the instrument that carries out quantitative PCR at microtiter plate.

In other examples, use

RT-PCR commercial measurement mRNA level.Can use the obtainable equipment of commercial sources to carry out RT-PCR.This system can comprise equipment (CCD) camera and the computing machine of thermal cycler, laser, electric charge coupling.In some instances, system on thermal cycler with 96 hole form amplified sample.During increasing, the fluorescence signal of induced with laser passes through the fiber optic cables real-time collecting in all 96 holes, and at CCD, detects.This system comprises for the operation instrument with for the software of analyzing data.

, for the impact that changes between minimum error and sample and sample, can use internal standard to carry out RT-PCR.Desirable internal standard is expressed with constant level between different tissues, and is not tested the impact of processing.The RNA that is usually used in the standardization gene expression pattern is the mRNA of house-keeping gene GAPDH, beta-actin and 18S rRNA.

Provided the originate step of the representative solution that quantitate gene expresses with fixing paraffin-embedded setup action RNA in the magazine article of various publication, comprise that mRNA separation, purifying, primer extend and increase (referring to people such as Godfrey, J.Mol.Diag.2:8491,2000; The people such as Specht, Am.J.Pathol.158:419-29,2001).In brief, representational method starts to cut the paraffin-embedded tumor tissues sample section that approximately 10 μ m are thick.Then extract RNA, and remove protein and DNA.Alternatively, from tumor sample or the direct isolation of RNA of other tissue samples.After analyzing RNA concentration, can comprise that if necessary RNA repairs and/or amplification step, and use gene specific promoter reverse transcription RNA, carry out subsequently RT-PCR and/or with nucleic acid array, hybridize.

In optional embodiment, known in the art for quantitatively the common method of sample mrna expression can use together with colon provided herein is signed.These class methods include but not limited to northern trace and in situ hybridization (Parker﹠amp; Barnes, Methods in Molecular Biology106:247283 (1999)); The RNA enzyme protection is measured (Hod, Biotechniques13:852 854 (1992)).Alternatively, can adopt the antibody that can identify specific duplex, described duplex comprises DNA duplex, RNA duplex and DNA-RNA heteroduplex body or DNA-protein duplex.

The technology of other PCR-baseds comprises that for example difference shows (Liang and Pardee, Science257:967971 (1992)); AFLP (iAFLP) (people such as Kawamoto, Genome Res.12:1305 1312 (1999)); BeadArray ^TMTechnology (Illumina, San Diego, Calif.; The people such as Oliphant, Discovery of Markers for Disease (Supplement to Biotechniques), in June, 2002; The people such as Ferguson, Analytical Chemistry72:5618 (2000)); BeadsArray (BADGE) for detection of gene expression, Fast Measurement (the people such as Yang in gene expression, Genome Res.11:1888 1898 (2001)) middle microballoon (the Luminex Corp. that uses the obtainable Luminex100LabMAP system of commercial sources and multi-coloured codes, Austin, Tex.); Competitive PCR and MassARRAY (people such as Oeth, 2004, SEQUONOME Application Note); Cover express spectra (HiCEP) analysis people such as (, Nucl.Acids.Res.31 (16) e94 (2003)) Fukumura with height.

The primer that is used for amplification be selected as increasing unique section of gene of interest (for example listed gene of table 1, table and table 6).But the primer commercial sources that can be used for these obtains, perhaps can according to known method for example use with

The sequences Design of these genes that obtain and synthetic.

Optional quantitative nucleic acid amplification program description is in U.S. Patent number 5,219,727.In this program, measure the amount of target sequence in sample by while amplified target sequence and internal standard nucleic acid segment.From the amount of the DNA of the amplification of each section determined and with the typical curve comparison to determine the amount of the target nucleic acid section that exists in sample before amplification.

In some instances, use microarray technology to identify or confirm gene expression.Therefore, use microarray technology, can measure express spectra in fresh or paraffin-embedded tumor tissues.In the method, interested colon cancer is signed nucleotide sequence (comprising cDNA and oligonucleotides) by bed board or is arranged in the microchip substrate.Then, the sequence of arrangement is hybridized with the nucleic acid that separates from interested cell or tissue (for example cDNA or mRNA).In the RT-PCR method, the source of mRNA is normally from people's tumour and the optional total RNA that separates with normal structure or clone from corresponding cancerous lung tissue.

In the particular of microarray technology, the embolus of cDNA clone's pcr amplification is applied to substrate with closely spaced array.In some instances, array comprises at least two special probes of colon cancer signature gene in his-and-hers watches 1,2 and 6.Microarray nucleic acid is fit to hybridize under stringent condition.Can produce fluorescently-labeled cDNA probe by the fluorescent nucleotide of incorporating into via the reverse transcription of the RNA that extracts from tissue of interest.Be applied to each the some specific hybrid of DNA on the cDNA probe of mark of chip and array.After strict washing is with the probe of removing non-specific binding, by confocal laser microexamination or by another detection method CCD camera scanning chip for example.The corresponding mRNA abundance of quantitative permission assessment of the hybridization of the element of each arrangement.Use dual base color fluorescence, the cDNA probe of the separate marking that produces from two RNA sources can be in couples and hybridization array.Therefore, measure simultaneously and specify the relative abundance of the transcript of gene from two sources corresponding to each.The hybridization scale of miniaturization provides the convenience of colon cancer signature gene expression pattern in his-and-hers watches 1,2 and 6 and has estimated fast.Can carry out microarray analysis, for example Affymetrix by the obtainable equipment of commercial sources according to manufacturer's scheme

Technology (Affymetrix, Santa Clara, CA) or Agilent microarray technology (Agilent Technologies, Santa Clara, CA) provide.

3. the additional method of gene expression analysis

The continuous analysis (SAGE) of gene expression is when allowing the lots of genes transcript and the other method of quantitative test, does not need to provide the individual hybridization probe of each transcript.At first, produce the short sequence label (approximately 10-14 base-pair) that contains the information that is enough to unique evaluation transcript, condition is that described label is available from the unique location in each transcript.Then, many transcripts are joined together to form the continuous molecule of the length that can be sequenced, and therefore disclose simultaneously the identity of a plurality of labels.Abundance that can be by measuring individual tag and identify in response to the gene of each label estimate quantitatively any transcript colony expression pattern (referring to for example, the people such as Velculescu, Science270:484-7,1995; With the people such as Velculescu, Cell88:243-51,1997).

In situ hybridization (ISH) is for detection of the other method with more interested gene expression.ISH apply and the nucleic acid hybridization technique of extrapolating to unicellular level, and with cytochemistry, immunocytochemistry and immunohistochemistry technology's combination, allow the form of cell sign thing and keeping of evaluation to be maintained and that identify, and allow sequence to be positioned to intragroup specific cells, for example tissue and blood sample.ISH uses complementary nucleic acid with the part of position tissue or section (original position) if or the enough little hybridization of a type one or more specific nucleic acid sequences of (the whole ISH of installation) in whole tissue of tissue.Can use the expression pattern in RNA ISH mensuration tissue, for example expression of cancer survival factors related gene.

Process the sample cell or tissue to increase its permeability to allow probe, for example cancer survival factors related gene specific probe, enter cell.This probe is added into the cell of processing, allows to hybridize under associated temperature, and washes excess probe off.The mark complementary probe, make and can for example with radioautograph, fluorescence microscopy or immunoassays, measure position and the amount of the probe in tissue.Sample can be any sample described herein, for example non-tumor sample or mammary gland or lung neoplasm sample.Because the sequence of interested cancer survival factors related gene is known, correspondingly designing probe, make probe specificity in conjunction with interested gene.

Original position PCR is the amplification of the PCR-based of target nucleic acid sequence before ISH., for the detection of RNA, introduce the interior reverse transcription step of cell to produce complementary DNA from the RNA template before PCR in position.This has realized the detection of low copy RNA sequence.

Before PCR, the cell or tissue sample is fixed and is penetrating to keep form and to allow PCR reagent to arrive sequence in cell to be amplified in position.Next in the intact cell of keeping in suspension or the direct centrifugal pcr amplification that carries out target sequence in sheet articles or histotomy that is coated with on microslide.In method before, use conventional thermal cycler to make the fixed cell that suspends in the PCR reaction mixture carry out thermal cycle.After PCR, the cell centrifugation smear, to microslide, is wherein passed through PCR product in ISH or immunohistochemistry showed cell.By the overlapping original position PCR that carries out on microslide of PCR potpourri with under sample and cover glass, it is then sealed to prevent the evaporation of reaction mixture.The heat block top of the thermal cycler by microslide being placed directly in routine or particular design or by using heat-circulation oven to realize thermal cycle.

Generally by one of following two kinds of different technology, realize the detection of in cell PCR product: the indirect in situ PCR of the ISH by using PCR product specific probe, perhaps the nucleotide by the direct-detection mark (for example foxalin-11-dUTP, fluorescein-dUTP, ³The Direct in situ PCR without ISH of H-CTP or biotin-16-dUTP), the nucleotide of described mark is incorporated in the PCR product during thermal cycle.

In some embodiments of detection method, can also estimate the expression of one or more " house keeper " genes or " internal contrast ".These terms comprise its existence can realize any composition of evaluation of cancer survival factors related gene (or protein) level or gene that the overall situation is expressed (or protein, as discussed below).This evaluation comprises the control that the mensuration of main assembly level of genetic transcription and RNA (or protein) change in reclaiming.

The disclosure is also by the further example explanation of following non-limiting example.

Embodiment

Embodiment 1

The present embodiment has been described generation and the checking of the exemplary forecasting tool that uses method disclosed herein and reagent classification colon cancer sample.The present embodiment comprises by people such as Kennedy, J.Clin.Oncol., and 29 (35) 4620-4626, the disclosed material of inventor of subject technology in 2011, described document integral body is by reference incorporated this paper clearly into.

Develop colorectal cancer and transcribed group focusing research array (Colorectal Cancer DSA ^TM(Almac Diagnostics, N.Ireland; It may reside in WWW almac-diagnostics.com)), can send accurately expression data people such as (, J Clin.Oncol.24:3519,2006) Johnston from the derivative RNA of FFPE.

Colorectal Cancer DSA ^TMResearch tool contains 61,528 probe groups and 52,306 transcripts that are proved to be at colon cancer and normal tissue expression of encoding.Use BLAST to analyze, for the mankind of U.S. biotechnology information center (NCBI) reference sequences (RefSeq) RNA database (it may reside in WWW ncbi.nlm.nih.gov/RefSeq/), compared Colorectal Cancer DSA ^TMResearch tool, 21,968 (42%) transcripts are present in mankind RefSeq database, and 26,676 (51%) transcripts are not present in mankind RefSeq database.And 7% content represents the antisense transcript of the expression of note gene.(people such as Johnston, J.Clin.Oncol.24:3519,2006; The people such as Pruitt, Nucleic Acids Research33:D501-D504,2005).In addition, the Colorectal Cancer DSA that compares with main general array ^TMThe analysis of probe level, outstanding about 20,000 (40%) transcripts are not included on main general microarray platform (Affymetrix), and are Colorectal Cancer DSA ^TMUnique.Therefore, Colorectal Cancer DSA ^TMResearch tool comprises does not also have transcript available in the gene expression research of carrying out so far.Finally, because be used for the Colorectal Cancer DSA of design ^TMThe transcript message part produce by the high-flux sequence method, may produce than the probe of the more close transcript 3 ' end that contains on other general microarraies.Relevant disease specificity content and based on the combination of 3 ' probe design, produced and can carry out from the derivative RNA of FFPE the unique product of powerful analysis of spectrum.

The purpose of this research is to estimate Colorectal Cancer DSA ^TMThe purposes of research array, produce and verify and can exactly II phase colorectal cancer patients be categorized as the prognosis gene signature of low or high risk recurrence after operation independently with the derivative tumour material of FFPE.The II phase colon cancer of using in the present embodiment is AJCC T3 or negative (NO) non-transfer (MO) colon cancer of T4 knot.

Method

Sample is selected

Collect sample with following criterion of acceptability respectively: II phase adenocarcinoma of colon does not only have the evidence of residual disease; 45 years old age during first operation or larger patient; Arrive 6 or multizone lymph node more; There is minimum 50% tumour cell in histotomy; The family history that there is no colon cancer; Perform the operation in 1 year there is no operation before or postoperative cancer therapy (but the therapy that gives after recurrence is acceptable); With follow up a case by regular visits to for the low-risk patient minimum patient of 5 years.Low-risk patient is defined in first operation does not have those of cancer return in 5 years.High-risk patient is defined as first operation and has those of cancer return of transfer in 5 years.Patient with local disease recurrence is excluded, because this recurrence may be Local residues disease but not the result of metastatic tumour after operation.Sample is collected at the Cong12Ge center.All sample experience virologists independently histopathology look back.Data group and supervision, epidemiology and the comparison of net result database have represented to guarantee it the general groups of suffering from II phase colon cancer.Crucial patient and tumour feature provide (referring to Fig. 6) at table 3.

Gene expression profile from the FFPE tissue

Use Roche high pure rna paraffin kit (Roche, Basel, Switzerland) to extract total RNA from the FFPE tumor sample.Use Nugen

FFPE System v2 and Nugen

The cDNA target of cDNA Biotin Module v2 combination preparation amplification, and according to manufacturer's instructions, carry out.Carry out hybridization, washing, dyeing and the scanning of the mark cDNA of fragmentation according to the Affymetrix scheme of standard.3.0 the mark cDNA to 3.5 μ g fragmentations goes up and Colorectal Cancer DSA at Affymetrix7G scanner (Affymetrix, Santa Clara, CA) ^TMMicroarray (Almac, Craigavon, United Kingdom) hybridization.Used the shape sample scheduling strategy, it comprises sample is divided into batch, and described batch according to operator, reagent and material charge number and target is clinical and the sample properties factor is carried out randomization.Applied quality control standard, and between low-risk and excessive risk sample balance biology and technical factor.It is carried out with the minimization system sexual deviation and makes any residual technology deviation diffuse into technique variation.

Sorter model is identified

Model development starts from 5,014 probe groups, and it is accredited as stable and/or has the tissue of suitable longitudinal stability with the difference degraded of avoiding probe groups under FFPE is fixing.Use subsequently the generation of signing of offset minimum binary classification, wherein select to eliminate based on recursive feature the key character of (RFE) during 10 times of 5 times of cross validations repeat.All aspects of model development suitably are nested in cross validation, comprise inceptive filtering removing 50% probe groups with minimum variation and intensity, based on average (RefRMA) standardization of the powerful multi-chip of reference and summarize and abandon during each the repetition RFE of the most unessential 10% probe groups.Measure by the characteristic length have maximum average area (AUC) under receiver operating-characteristic curve according to cross validation the feature sum that final mask comprises.Based on the sensitivity from the cross validation training data and specific add and maximal value (minimum value (Youden, Cancer3:32-35,1950) of Youden J statistics is selected from the threshold value of the dichotomy of each model prediction., in the situation that have a plurality of threshold values of identical performance substantially, use the danger from Cox ratio hazards regression to be beneficial to higher HR value than (HR) as play-off competition.

The reprography of the colorectal cancer cell system (HCT116) that embeds in the FFPE by prediction and the synchronic analysis of clinical sample is estimated the precision of prediction.The commercial measurement of the repetition of this sample is not included in model development, but by all 50 the cross validation training subgroups as the independent test group, predicts, purpose is to select the model with high duplication and repeatability.In addition, arrange test, wherein real class label is reorganized 100 times at random, follows and develops by complete model.This is carried out to assess and can accidentally expects which type of classification performance from the data group with these features, and discloses any deviation in the signature generating routine.

Use single argument and multivariate Cox ratio hazards regression to estimate the independence of final mask in known clinical factor environment.The input of using is the class label of two minutes of prediction, together with the lymph node number of tumour stage, patient tumors grade, knub position, patient age, Gender, mucus/non-mucus hypotype and retrieval.Microsatellite instability is not included as factor, because this information is not that most of sample is obtainable.Use is carried out the enrichment of Gene Ontology note and Gene Ontology bioprocess and molecular function based on the intrinsic developing instrument of gene in final signature.Determine the functional category of the gene of significant enrichment with the hypergeometric distribution with the correction of false discovery rate multiple check.By using Ingenuity Pathway Analysis (Ingenuity Systems, Redwood City, CA) to produce path analysis.

The balance of sample, randomization and quality control (QC)

Target group: be used for the suitable general groups character that reflects from SEER and CRUK database of colony that training is measured.Consider following character:

Sex.In II phase colony sex popular be about 50-60% male sex (be 56% in Britain, and in the U.S., be 57%).

Knub position (far-end/near-end).Popular in II phase colony is about 55%-65% near-end and 35%-45% far-end.

Patient age.According to the SEER cancer statistics summary coton and rectal part of the NCI of 2001-2005,0.1% patient was diagnosed below 20 years old; 1.0% 20 and 34 years old between; 3.7% 35 and 44 years old between; 11.6% 45 and 54 years old between; 18.3% 55 and 64 years old between; 25.1% 65 and 74 years old between; 28.2% 75 and 84 years old between, and 12.2% more than 85 years old.

Without the recurrence survival rate.It is reported in II phase colony that without the recurrence survival rate be 13%-22% people such as (, European Journal of Cancer, 2006) Gattaj, and according to the SEER database, be approximately 30%.

Pre-equilibration: carry out pre-equilibration, make the sample group that is proposed for hybridization be balanced with respect to the clinical co-variation amount of selecting, the above general groups statistics that provides is provided simultaneously.This gets rid of without recurring survival, and it is had a mind to enrichment to increase the ability of biomarker discovery.The training group does not contain any sample that has event after 5 years, and this is not the constraint in the checking group.Do not use 5 years afterwards the sample of recurrence sign the principle of generations (that is, in the training group) be for fear of when carrying out the biomarker discovery with extra inhomogeneity introducing sample colony.

The fundamental purpose of equilibrium code is to reduce terminal (as the high/low risk of binary variable representative) and the following factors related (if there is) between any one.Any association between these factors and high/low risk terminal will be introduced obscuring of the clinical efficacy that can limit mensuration.Make 603 colorectum samples experience pre-equilibrations to reduce the strong association between any one of prognosis and following factor: sex; Knub position in intestines; Patient age; Contribution centre; The FFPE piece age (date of surgery); Tumour content; With the RNA quality.

Check continuous parameter with Kolmogorov-Smirnoff, and with Chi-square Test, carry out the inspection-classification parameter.Require the p value of all parameters 〉=0.4, to realize balance.504 samples keep (335 low-risks and 169 excessive risks) after balance, and are suggested to array analysis.

Sample randomization during array analysis: carrying out the sample randomization to avoid obscuring between known technology and biodyne, is mainly interested terminal (prognosis).In this research operator, hybridization-washing-dyeing (HWS) kit lot number, array lot number and array batch are considered together with contribution centre and prognosis.At first sample is dissolved array batch at random, make each array batch have prognosis and the contribution centre of same ratio.Then according to availability, the operator is dispensed to each array batch.Distribute the HWS kit then for each array batch, guarantee that each operator uses the same ratio of each kit.The array lot number is assigned to each array batch, guarantees their mean allocation in array batch.

The quality control of training data: mainly based on the value in the Affymetrix RPT document that contains various quality correlation parameters, the QC program is applied to the array that obtains., for all samples, calculate limit value: % based on the visual examination of the distribution of each parameter and call out at present (requiring 〉=20%); Identify that the image processed goods has the array of obvious stain with removal; From the T based on Q residual sum Hotelling ²Key component analysis (PCA) detect exceptional value.

Determine with the evaluation of sex gene whether the expression of observing mates sex known in clinical information.

Also considered following Affymetrix mass parameter during the visual examination that distributes; Extensively be categorized as following: the RNA quality; Xin Haozhiliang ﹠amp; Detection of call; Bei Jing ﹠amp; Noise; With the background homogeney.

Amount to 319 colorectum samples and passed through the QC program.Due to the heterogeneity that the PRELIMINARY RESULTS hint is introduced by rectal sample, remove rectal sample to form 249 only groups of colon, it is proposed for finally (after QC) balance.

Final QC back balance: use 249 the colon samples of the principle balance identical with initial pre-equilibration by QC, increase the present call distribution of % similar standard (this information can obtain after only hybridizing) in low-risk and excessive risk group.Keep the group of final 215 samples after QC and balance.

Final colon group with 215 samples is compared and is had following character with known population distribution: sex: 53% male sex (50-60% in colony); Knub position (far-end/near-end): 62% near-end (55-65% near-end in colony); Patient age: immediately following the continuous distribution of colony; Recur survival rate with nothing: 34% poor prognosis (excessive risk).Compare enrichment intentionally with about 15-20% colony.

The quality contrast of checking group and following sample group: it is important step that the training group is used customization QC program, and purpose is to promote from quality data group identification of organism mark.Yet for the predict future sample, must be with the base application QC of next sample.And the QC program can not to the data group and wherein to produce the system of data too special.For this purpose, use 40 samples that copy in two systems and scanner to estimate separately, to identify QC parameter stable in system.AvgSigA parameter (average signal of absent probe groups) is confirmed as parameter the most stable in different system, and is therefore the best candidate of system independence QC program.For this parameter, higher value means lower quality, and lower value means higher quality.The strong negative correlation of the present calling parameter of % of the main QC parameter that the QC parameter that AvgSigA value and conduct are commonly used and training group are used.Come the low acceptance value that the % of self-training group calls out at present to be set as 20%, it is roughly corresponding to 43 upper acceptance value of the AvgSigA parameter of this data group., in order to adapt to younger FFPE sample, determine AvgSigA not to be introduced low threshold value (it will allow to comprise the better quality sample).Therefore, the scope that finally comprises that draws from this research is AvgSigA≤43, and it is the QC tolerance that is applied to the individual authentication group, and is to be applied to the QC of following sample.

Evaluation stable probe groups in the FFPE piece age: generally acknowledge that the mRNA transcript may and degrade to varying level with the different rates degraded in the FFPE sample, it may cause from the signature of the old material production that plays a role not as to fresh FFPE material expection.Therefore, carry out two independently longitudinal research to identify probe groups stable in the FFPE piece age.In the first research, 9 FFPE pieces are distinguished and continuously by DNA microarray 7 time point analyses in 16 time-of-week sections after fixing.These samples by in the one-year age section with three June interval the second longitudinal research supplement, wherein 8 FFPE piece scopes from 6 months to 4 years old the age, it is distinguished continuously and, by the DNA microarray analysis, is obtained 113 individual specimen for analysis.5014 transcripts that do not experience further degraded in time or with suitable speed, decompose after fixing have been identified.The list of this probe is used for subsequently signature and produces.Provide the independent manuscript of this research details in preparation.

The accuracy of estimation model sorter between development stage: when being used for test environment, sorter is the importance of measuring from the consistent ability that produces identical output of reprography.For this purpose, one group of 39 reference sample that is the reprography of (HCT114) as identical colorectal cancer cell are hybridized together with clinical sample.During model development, this group is predicted to be the external inspection group during cross validation, with the relative variance in per step in the estimation model performance history.There is no information sharing between training group and 39 sample reference group during cross validation.Calculate the standard deviation of with the signature of prediction, marking and be shown as the mean value with 95% confidence limit.For long signature, variability is low, and then along with the feature selecting program increases gradually, it is also reflected in the low accuracy (AUC) of shorter signature.Under the signature length (634 probe groups) of selecting, models show high precision and accuracy.

The arrangement analysis of classification performance: carry out arrangement analysis and can accidentally expect what classification performance from the data group with similar quality to estimate.This also repeats subsequently whole model development process (having filtration, standardization, feature selecting and classification) by the true class label of random reorganization (that is, true prognosis) and carries out.The signature performance significantly be better than longer signature length and particularly wherein the probe groups number be 634 selected person's chance.In addition, arranging test has disclosed for the data group of exploitation sorter and/or any potential deviation of method.Intermediate value AUC in random tags is 0.5, the indication chance, and it confirms there is no obvious deviation in the program of using.

Result

Sign from the prognosis of FFPE organization development II phase colon cancer.5 years without disease, survive as the main terminal of this research.After being applied to the primary data group at the balance clinical factor and with quality control standard, identified 215 patients' (142 low-risks and 73 high-risk patients) training group.Carrying out 50% variance-intensity filtrations, RefRMA standardization, RFE feature selecting and offset minimum binary under 10 times of five times of cross validations be used for estimating classification performance repeat classifies.634 transcript signatures of cross validation indication are that the prognosis classification is best.Generation has 0.68 (receiver operating-characteristic curve of the AUC of P＜.001), the remarkable association (Fig. 3 A) between the scoring of indication signature and prognosis.The AUC that observes is significantly higher than random in arrangement analysis, and shows low variance the evaluation from the reprography prediction.Having set up 0.465 threshold value of the dichotomy of signature prediction scoring from Youden J statistics, having produced HR (P＜.001 of 2.62; Fig. 3 B).Table 4 contains the classification performance general introduction of signing and producing during cross validation.

The classification performance of table 4. training and individual authentication group

95%CI be from cross validation (training group) or have 1,000 repetition the boots method of pulling out (checking group) ± 2 standard deviations; When calculating NPV and PPV, used respectively 80% and 20% priori value.Threshold value t=0.465 is used for the dichotomy of signature scoring.Abbreviation: AUC, area under receiver operating-characteristic curve; HR, dangerous ratio; NPV, negative predictive value (feminine gender is low-risk); PPV, positive predictive value (positive is excessive risk).

The individual authentication of II phase colon cancer prognosis signature: the prognosis signature is applied to and uses the threshold value of identifying in the training group to mark for 144 patients' (85 low-risk patients and 59 high-risk patients) of recurring enrichment individual authentication group.Carry out separately sample analysis, and in time after a while to the training group.(the palindromia (Fig. 4 and table 4) of the HR of P＜.001) has been predicted in the excessive risk group and had 2.53 to signature.(the cancer associated death (Fig. 5) of P＜.0084) has also been predicted in the excessive risk group and had 2.21 to signature.

Signature described herein is conducive to extensive authentication policy based on the retrospective analysis in existing FFPE tumour storehouse from the fact of the derivative tumour developing material of FFPE.

Dangerous than be accredited as event danger in high risk II phase colorectal cancer patients or chance by sorter be accredited as the expression of the ratio of event danger in low-risk patient by sorter.In rear 5 years of operation, those that have a poor prognosis with predicted are compared, and are predicted to be the group with good prognosis and have significantly lower recurrence probability.Negative predictive value is the ratio through the patient with negative test result (prediction is negative) of correct diagnosis.In prognosis arranged, NPV relied on the popular of palindromia.Positive predictive value is the ratio through the patient of the positive test result of having of correct diagnosis (Predict masculine gender).In prognosis arranged, PPV relied on the popular of palindromia.Colony based on 20% poor prognosis sample is popular, and this will mean in 5 years, and the patient with poor prognosis of prediction has 33% recurrence probability, and the patient with good prognosis of prediction has 13% recurrence probability.

Assess signature independence from known Prognostic Factors: measure for useful prognosis, its known Prognostic Factors that must be independent of clinical use carries out.Therefore, assessed the independence (table 5) of measuring in single argument and multivariable analysis.

The comparison of table 5. transcript label and individual authentication group Plays pathology parameter

Used the Cox ratio hazards regression with P value of checking from log-likelihood ratio to carry out single argument and multivariable analysis.For the tumour grade, grade 1 is as the reference point of calculating HR.The footing order of patient age and retrieval is as factor is analyzed continuously.The explanation of the HR of patient age is 1 years old increase risk that changes, and correspondingly, and the explanation of the footing purpose HR of retrieval is the risk of the increase that increases of the knot of a retrieval.Abbreviation: HR, dangerous ratio.

((P＜.001) is significant in analyzing, and illustrate signs provides the information of the prognosis except the normal risk factor for P＜.001) and multivariate in single argument in the prediction of prognosis.And, invade to assess the independence of signature by increasing lymph vessels in the sample recording it (in the checking group in 144 samples 100).Signature is with single argument (P＜.001) and multivariable analysis (P＜.001) independently carry out.

The functional analysis of gene in the prognosis signature: next, ask whether mensuration has detected the known bioprocess relevant to the colon cancer recurrence.Using Ingenuity Pathway Analysis to analyze 634 probe groups, and identified the list in statistically significant path, is wherein the conduction of IGF-1 signal the most significantly.

Discuss

Discuss as this paper, developed and identified after II phase Operation of Colon Carcinoma the mensuration based on DNA microarray that is in the patient of higher risk of recurrence.Particularly, this signature has been identified in the individual authentication group excessive risk formation of the cancer associated death HR with 2.53 recurrence HR and 2.21.The checking of using the prognosis of group fully separately to measure is to avoid the performance of training group signature to over-evaluate necessary.2.53 recurrence HR advantageously be used for simultaneously making clinical decision usually have about 1.5 or the histological of less HR compare.And signature does not need individual the explanation, and can provide than the more standardized method of conventional organization Pathologic factors.Importantly, this mensuration is carried out the FFPE tissue, and therefore easily is applied to present medical practice.

Although disclose several prognosis based on DNA microarray test in several cancer types, a kind of being introduced in clinical practice only, and so far neither one for colon cancer.This may be due to two principal elements.At first, from fresh or freezing tissue, many signatures have been developed.The second, unsuitable research method has caused the failure with the validation test of independent data group.

, about the use of freezing tissue sample,, although this types of organization provides good microarray data, from the test that this tissue produces, can not fully carry out the FFPE tissue.This can produce difficulty in collection is enough to develop the sample of testing with the individual authentication prognosis.Sample in addition, needs the variation of clinical practice based on the enforcement of the mensuration of flesh tissue, because need to be collected when operation.

FFPE is the standard of tumour filing, and has existed for many tumours storehouse of measuring exploitation.Importantly, based on the exploitation of the mensuration of FFPE and the variation that clinical implementation does not need sample collection and processing.

Develop disclosure method so that FFPE is organized effect, but used the DNA microarray platform, thereby greatly increased the number of detectable mRNA transcript and bioprocess with respect to the quantitative polyase chain reaction technology., owing to having used the FFPE material with microarray platform, need to consider several method tissue.Formalin fixedly causes the degraded of mRNA transcript by RNA and protein cross.Great majority should occur at once in degraded, but some transcripts continue degraded in time.The DNA microarray platform that is used for research has the probe groups that designs for mRNA transcript 3 ' end, to strengthen the ability of the transcript that detects degraded.In addition, analyzed in time the colon cancer sample of independent group, used and guarantee that we do not incorporate the probe groups that detects unstable or the mRNA transcript that difference the is stable part as signature into.

The predicted value of signature is higher than also surpassing the known clinical co-variation amount of prognosis.This performance can promote the initial balance of the prognosis of the biology that carries out for the part as setting up suitable training group and technical factor greatly.The biodyne of considering comprises known Prognostic Factors, and for example pT stage and grade, and other non-Prognostic Factors that may affect gene expression, comprise knub position, patient age and sex.Technical factor, for example also balance between height and low-risk sample in the training group of FFPE piece age and contribution centre.In addition, carry out the randomization of operator and kit to avoid obscuring between technical factor and known clinical factor.This has at utmost reduced to measure and depends on the operator or depend on risk from the use of the use of the sample at specific center or particular batch reagent.Be independent of known Prognostic Factors because measure to be developed to, we believe to develop and incorporate several factors into to produce the multi-parameters test of pre-back pointer even more accurately.

The functional analysis of gene signature discloses, and the conduction of IGF-1 signal, the conduction of TGF-signal beta and the conduction of HMGB1 signal belong to the most important path of evaluation.All these before had been reported in colon cancer by promoting tumor growth, intrusion and transfer and preventing that apoptosis from giving poor prognosis.In a word, herein disclosed is the empirical tests of II phase colon cancer of the tumor tissues that stores for FFPE and powerful prognosis DNA microarray signature.

Signature of the present disclosure can help the doctor in risk of recurrence and benefit from aspect the possibility of NACT and make wiser clinical decision (people such as Andre, Annals of Surgical Oncology13:887-898,2006; The people such as Diaz-Rubio, Clin.Transl.Oncol.7:3-11,2005; The people such as Monga, Ann.Surg.Oncol.13:1021-1134,2006; Sobrero, Lancet Oncol.7:515-516,2006).And many patients wonder their curability and Operative risk/benefit (people such as Gill, J.Clin.Oncol.22:1797-1806,2004; The people such as Kinney, Cancer91:57-65,2001; The people such as Carney, Ann.R.Coll.Surg.Engl.88:447-449,2006; Salkeld, Health Expect7:104-1014,2004).Can predict that patient's prognosis provides the better assessment that risk/benefit and therapy are selected for doctor and patient.The ability that provides personalized patient to nurse will be hopeful to provide for these patients survival and the quality of life of improvement.

In the past, many researchs have hinted that sample size is the main cause that lacks credible statistics evidence, and point out that larger test is that proof supplemental treatment benefit is necessary.Use the prognostic marker of checking, the gene signature that for example produces in this research, II phase patient can be divided into high and low-risk subgroup.The method can design by the clinical trial that absorbed those patients that are in high risk of recurrence help improve, and the benefit of the complementary therapy of therefore more may deriving.Therefore, Colorectal Cancer DSA ^TMCan be useful research tool, be used for the classification patient to be included in clinical testing, be used for making decisions about auxiliary and new supplemental treatment, and for the identification of new route or the molecule target of extra drug exploitation.

In table 6, the prognosis signature of report has been predicted the recurrence of II phase colon cancer exactly, and based on FFPE checking group independently, estimates.Overall accuracy for the recurrence prediction of this heterogeneous disease is essence.Colony based on 20% poor prognosis sample is popular, this means in 5 years, and the patient with poor prognosis of prediction has 33% recurrence probability, and the patient with good prognosis of prediction has 13% recurrence probability.One of major advantage of method is its express spectra based on the FFPE tissue at present, and it is the preferred storage method (Abramovitz Proteome Sci.4:5,2006) for most of available tissue banks.The RNA that extracts from the FFPE tissue samples often has shorter intermediate value length due to the modification of degrading and formalin is induced, and this makes general mensuration be difficult to detect.When definite colon cancer is transcribed group, adopt the sequence measurement based on 3 ', this is conducive to design the probe groups for 3 ' end of each transcript.The method is guaranteed much higher detection rates, and therefore optimised design from fresh food frozen and FFPE tissue samples, to detect the rna transcription thing.The result of this research shows, Almac Diagnostics Colorectal Cancer DSA ^TMResearch tool can produce biological meaningful and reproducible data from FFPE derived weave.

Embodiment 2

The prognosis of cancer

The present embodiment has been described and can be diagnosed with for prognosis the experimenter's of colon cancer ad hoc approach.Yet, it will be understood by those skilled in the art that the experimenter's who suffers from colon cancer prognosis can also successfully be provided with the method that departs from these ad hoc approach.

Tumor sample and adjacent non-tumor sample are available from the experimenter.For example use fine needle aspiration to obtain about 1-100 μ g tissue to each sample type.Use conventional method (for example using the commercial reagents box) from tumour and nonneoplastic tissue isolation of RNA and/or protein.

In an example, determine the prognosis of colon cancer tumour by microarray analysis or real-time quantitative PCR detection table 1,2 and/or 6 transcription things 2 or more expressions in the tumor sample available from the experimenter.For example, can utilize disclosed gene signature.With the relative expression's level in tumor sample and contrast (RNA that for example, from experimenter's adjacent nonneoplastic tissue, separates) relatively.In other cases, contrast is reference value, the relative quantity of this molecule that for example exists in the non-tumor sample available from one group of health volunteer or cancer experimenter.

, in view of the adaptable many possible embodiments of principle of the present disclosure, should be appreciated that the embodiment of example explanation.

Claims (67)

1. one kind is used for the method for diagnosis available from experimenter's sample colon cancer, and the method comprises:

Detection is available from the expression of at least two kinds of listed colon cancer associated nucleic acid molecules of table 6 in experimenter's the sample that comprises nucleic acid; With

The expression of more described at least two kinds of colon cancer associated nucleic acid molecules or from the contrast threshold value of the decision-making of its derivation scoring with the indication diagnosis of colon cancer, wherein at the expression of described threshold value homonymy or from the diagnosis of the decision-making scoring indication colon cancer of its derivation, thereby diagnosis is available from the colon cancer in described experimenter's described sample.

2. method according to claim 1, wherein said contrast threshold value comprises:

The threshold value that always the corresponding transcript of the listed colon cancer associated nucleic acid of table 6 molecule is derived in known colon cancer sample (or a plurality of sample), wherein at the expression of the threshold value homonymy as known colon cancer group or from the diagnosis of the decision-making scoring indication colon cancer of its derivation.

3. method that is used for classification colon cancer sample, the method comprises:

Detection is available from the expression of at least two kinds of listed colon cancer associated nucleic acid molecules of table 6 in experimenter's the sample that comprises nucleic acid; With

The expression of more described at least two kinds of colon cancer associated nucleic acid molecules or from the contrast threshold value of the decision-making of its derivation scoring with the known classification of indication, wherein at the described expression of described threshold value homonymy or allow the classification of described colon cancer sample from the decision-making scoring of its derivation.

4. method according to claim 3, wherein said contrast threshold value comprises:

The threshold value that always the corresponding transcript of the listed colon cancer associated nucleic acid of table 6 molecule is derived in the colon cancer sample (or a plurality of sample) of known classification, wherein at the described expression of the threshold value homonymy of the colon cancer sample as known classification (or a plurality of sample) or allow the classification of described colon cancer sample from the decision-making scoring of its derivation.

5. according to claim 3 or 4 described methods of any one, wherein said colon cancer sample is classified as I phase, II phase, III phase and IV phase.

6. the described method of any one according to claim 3-5, comprise that further selection will be for the effective treatment plan of the colon cancer of described classification.

7. method according to claim 6, wherein said treatment is excision, chemotherapy, radiation or its combination in any.

8. one kind is used for the method for prediction to the response of the treatment of colon cancer, and the method comprises:

Detection is available from the expression of at least two kinds of listed colon cancer associated nucleic acid molecules of table 6 in experimenter's the sample that comprises nucleic acid; With

The expression of more described at least two kinds of colon cancer associated nucleic acid molecules or from the contrast threshold value of the known response to treatment of the decision-making of its derivation scoring and indication, wherein at the described expression of described threshold value homonymy or from the similar response to treatment of decision-making scoring indication of its derivation, thereby prediction is to the response for the treatment of.

9. method according to claim 8, wherein said contrast threshold value comprises:

The threshold value that the corresponding transcript of the listed colon cancer associated nucleic acid of table 6 molecule is derived from the colon cancer sample (or a plurality of sample) from having the known response to treatment, wherein at the described expression of the threshold value homonymy of the colon cancer sample (or a plurality of sample) as having known response to treatment or from the similar response to treatment of decision-making scoring indication of its derivation, thereby prediction is to the response for the treatment of.

10. according to claim 8 or 9 described methods of any one, wherein said treatment is excision.

11. the described method of any one according to claim 8-10, wherein said treatment is chemotherapy and/or radiation.

12. the method for the long-term surviving of predicting the experimenter who suffers from colon cancer, the method comprises:

Detection is available from the expression of at least two kinds of listed colon cancer associated nucleic acid molecules of table 6 in experimenter's the sample that comprises nucleic acid; With

The expression of more described at least two kinds of colon cancer associated nucleic acid molecules or have the contrast threshold value of long-term surviving history from the decision-making of its derivation scoring and indication, wherein at the described expression of described threshold value homonymy or from the described experimenter's of decision-making scoring indication of its derivation long-term surviving, thus prediction experimenter's long-term surviving.

13. method according to claim 12, wherein said contrast threshold value comprises:

The threshold value that the corresponding transcript of the listed colon cancer associated nucleic acid of table 6 molecule is derived from the experimenter available from having long-term surviving history (or a plurality of experimenter's) colon cancer sample (or a plurality of sample), wherein in conduct, available from the described expression of the threshold value homonymy of the experimenter with long-term surviving history (or a plurality of experimenter's) colon cancer sample (or a plurality of sample) or from the decision-making of its derivation, mark and indicate described experimenter's long-term surviving, thus prediction experimenter's long-term surviving.

14. method according to claim 13, wherein long-term surviving comprises survival at least 5 years.

15. a method that is used for the recurrence of prediction experimenter colon cancer, the method comprises:

Detection is available from the expression of at least two kinds of listed colon cancer associated nucleic acid molecules of table 6 in experimenter's the sample that comprises nucleic acid; With

The expression of more described at least two kinds of colon cancer associated nucleic acid molecules or from the historical contrast threshold value of the decision-making of its derivation scoring and indication recurrence, wherein in described expression or the recurrence from the described experimenter of decision-making scoring indication of its derivation of described threshold value homonymy.

16. method according to claim 15, wherein said contrast threshold value comprises:

From from having, recurring the threshold value that historical colon cancer sample (or a plurality of sample), the corresponding transcript of the listed colon cancer associated nucleic acid of table 6 molecule is derived, the wherein recurrence among the described expression of the threshold value homonymy as the historical colon cancer sample (or a plurality of sample) of known recurrence or the described experimenters of indication that marks from the decision-making of its derivation.

17. a method for preparing the personalized colon cancer genome spectrum of experimenter, the method comprises:

Detection is available from the expression of at least two kinds of listed colon cancer associated nucleic acid molecules of table 6 in experimenter's the sample that comprises nucleic acid; With

Produce the report of summarizing the data that obtain by described gene expression analysis.

18. the described method of any one according to claim 1-17, wherein comprise from RNA and/or the cDNA of the rna transcription of the sample extraction of the colorectum tissue available from described experimenter available from described experimenter's nucleic acid.

19. method according to claim 18, wherein said sample is biopsy sample.

20. the according to claim 18 or 19 described methods of any one, wherein said sample are fixing and/or paraffin-embedded sample.

21. the described method of any one according to claim 1-20, wherein said expression carrys out standardization according to a kind of crt gene or multiple crt gene.

22. the described method of any one according to claim 1-21, wherein said expression is measured with PCR and/or based on the method for microarray.

23. the described method of any one according to claim 1-22, wherein the expression of the listed at least two kinds of colon cancer associated nucleic acid molecules of detection table 6 comprises the expression that detects MUM1 and SIGMAR1 transcript.

24. the described method of any one according to claim 1-23, wherein the expression of the listed at least two kinds of colon cancer associated nucleic acid molecules of detection table 6 comprises the expression that detects MUM1, SIGMAR1, ARSD, SULT1C2 and PPFIBP1 transcript.

25. the described method of any one according to claim 1-24, wherein the expression of the listed at least two kinds of colon cancer associated nucleic acid molecules of detection table 6 comprises the expression of the antisense sequences of the expression that detects ARSD, CXCL9, PCLO, SLC2A3, FCGBP, SLC2A14, SLC2A3, BCL9L transcript and MUC3A, OLFM4 and RNF39 transcript.

26. the described method of any one according to claim 1-25, wherein the expression of the listed at least two kinds of colon cancer associated nucleic acid molecules of detection table 6 comprises the expression of the listed transcript of detection table 1.

27. the described method of any one according to claim 1-26, wherein the expression of the listed at least two kinds of colon cancer associated nucleic acid molecules of detection table 6 comprises the expression of the listed transcript of detection table 2.

28. nucleic acid probe for detection of the colon cancer gene expression signature, described nucleic acid probe comprises length between the nucleic acid molecules between 20 and 40 nucleotide or substantially the nucleic acid molecules of length between 20 and 40 nucleotide, consists of, described nucleic acid molecules can with one of nucleotide sequence as shown in SEQ ID NO:1-636 or its complement specific hybrid.

29. nucleic acid probe according to claim 28, wherein said probe is mark.

30. nucleic acid probe according to claim 29, wherein said probe are radiolabeled, fluorescently-labeled, biotin labeled, enzyme labeling or chemical labeling.

31. probe groups for detection of the colon cancer gene expression signature, comprise two or more probes, wherein each probe comprises length between the nucleic acid molecules between 20 and 40 nucleotide or substantially the nucleic acid molecules of length between 20 and 40 nucleotide, consists of, described nucleic acid molecules can with one of nucleotide sequence as shown in SEQ ID NO:1-636 or its complement specific hybrid.

32. probe groups according to claim 31, wherein said probe is mark.

33. probe groups according to claim 32, wherein said probe are radiolabeled, fluorescently-labeled, biotin labeled, enzyme labeling or chemical labeling.

34. the described probe groups of any one according to claim 31-33, wherein said group of at least one probe that contains with each complementation of the listed transcript of table 6.

35. probe groups according to claim 34, comprise at least one probe of each transcript complementation in the subgroup with the listed transcript of table 6, wherein said subgroup comprises at least 1%, 5%, 10%, 25%, 50%, 75% or 95% of table 6 transcription thing.

36. the device for detection of the colon cancer gene expression signature, this device comprises the nucleic acid array that comprises the described probe groups of any one according to claim 31-35.

37. one kind is used for amplification colon cancer nucleic acid gene and expresses the primer pair of signature, comprising:

Length is the forward primer of 15 to 40 nucleotide, and it comprises and any one of nucleotide sequence as shown in SEQ ID NO:1-636 or the nucleotide sequence of its complement specific hybrid; With

Length is the reverse primer of 15 to 40 nucleotide, and it comprises and any one of nucleotide sequence as shown in SEQ ID NO:1-636 or the nucleotide sequence of its complement specific hybrid, and wherein said primer sets can instruct the amplification of described nucleic acid.

38. one kind is used for amplification colon cancer nucleic acid gene and expresses the primer pair group of signature, comprises any one at least two primer pairs of the nucleotide sequence as shown in SEQ ID NO:1-636 that is suitable for increasing.

39. described primer pair group according to claim 38, wherein said primer pair group comprises the primer pair of the subgroup of the nucleotide sequence as shown in SEQ ID NO:1-636 that is suitable for increasing, and wherein said subgroup comprises at least 1%, 5%, 10%, 25%, 50%, 75% or 95% of nucleotide sequence as shown in SEQ ID NO:1-636.

40. the method for the preparation of the gene expression profile of indication colon cancer prognosis, the method comprises:

Detection comprises the sample of the RNA that separates from the colon cancer sample expression less than 1000 transcripts, and wherein listed at least 50 transcripts of table 6 are detected.

41. described method according to claim 40, wherein 400 to 800 transcript expressions are detected.

42. according to claim 40 or 41 described methods, wherein 500 to 700 transcript expressions are detected.

43. the described method of any one according to claim 40-42 is wherein detected from least 100 transcripts of table 6.

44. the described method of any one according to claim 40-43 is wherein detected from least 200 transcripts of table 6.

45. the described method of any one according to claim 40-44 is wherein detected from least 300 transcripts of table 6.

46. the described method of any one according to claim 40-45 is wherein detected from least 400 transcripts of table 6.

47. the described method of any one according to claim 40-46 is wherein detected from least 500 transcripts of table 6.

48. the described method of any one according to claim 40-47 is wherein detected from least 600 transcripts of table 6.

49. the described method of any one according to claim 40-48, wherein all listed transcripts of table 6 are detected.

50. the described method of any one according to claim 40-49, wherein the listed transcript of table 6 comprises the transcript that table 1 is listed.

51. the described method of any one according to claim 40-50, wherein said colon cancer sample are the fixing paraffin-embedded tissue samples of formalin.

52. the described method of any one according to claim 40-51 also comprises: for the expression of transcript listed in the respective horizontal of excessive risk and low-risk patient colony or scoring his-and-hers watches 6 or from the decision-making of its derivation, mark.

53. 2 described methods, also comprise the selection NACT according to claim 5, wherein said patient is defined in described excessive risk group.

54. a method that is used for the prognosis colon cancer, the method comprises:

The gene expression profile for preparing the colon cancer sample of the RNA that comprises separation; With

Based on the expression of listed at least 50 transcripts of table 6 or from the decision-making scoring of its derivation with described specimen classification in low-risk or excessive risk group.

55. 4 described methods, wherein detect the level less than 1000 transcripts in described gene expression profile according to claim 5.

56. 4 or 55 described methods, wherein detect the level less than 800 transcripts in described gene expression profile according to claim 5.

57. the described method of any one in 4-56, wherein detect the level less than 700 transcripts in described gene expression profile according to claim 5.

58. the described method of any one in 4-57 according to claim 5, wherein said sample is classified based on the expression of 100 transcripts from table 6 at least.

59. the described method of any one in 4-58 according to claim 5, wherein said sample is classified based on the expression of 200 transcripts from table 6 at least.

60. the described method of any one in 4-59 according to claim 5, wherein said sample is classified based on the expression of 300 transcripts from table 6 at least.

61. the described method of any one in 4-60 according to claim 5, wherein said sample is classified based on the expression of 400 transcripts from table 6 at least.

62. the described method of any one in 4-61 according to claim 5, wherein said sample is classified based on the expression of 500 transcripts from table 6 at least.

63. the described method of any one in 4-62 according to claim 5, wherein said sample is classified based on the expression of 600 transcripts from table 6 at least.

64. the described method of any one in 4-63 according to claim 5, wherein the expression of listed all transcripts of table 6 is used for described sample is classified.

65. the described method of any one in 4-64 according to claim 5, wherein the listed transcript of table 6 comprises the transcript that table 1 is listed.

66. the described method of any one in 4-65 according to claim 5, wherein said colon cancer sample are the fixing paraffin-embedded tissue samples of formalin.

67. the described method of any one in 4-66, further comprise and select NACT when described patient be defined in described excessive risk group according to claim 5.

Images